P-591 Cellular heterogeneity in the preovulatory follicle correlates with response to stimulation with recombinant FSH and affects somatic cell gene expression levels

Abstract

Study question What is the difference in the composition and gene expression of pre-ovulatory follicular somatic cells between normo- and hyporesponder patients to rFSH stimulation? Summary answer Gene expression changes in the somatic cells of the preovulatory follicles of hyporesponder patients is partly attributable to distinct cell populations. What is known already Some women undergoing IVF respond suboptimally to the preceding ovarian stimulation without an indication of poor ovarian insufficiency nor advanced age. The gene expression levels in the somatic cells of their pre-ovulatory follicles have not been determined. However, the somatic cells play key roles in the transfer of gonadotrophin signals and in steroidogenesis. In the preovulatory follicle, being >20 mm in size, the somatic cells are exposed to various molecular gradients that drive their differentiation. Due to advancements in sequencing technologies the importance of somatic cell sub-populations in the etiologies of ovarian infertility can now be investigated by bioinformatic methods. Study design, size, duration Consecutive IVF patients <41 years of age were recruited at Nova Vita Clinic, Estonia, undergoing the antagonist stimulation protocol with rFSH administration. Ovarian puncture was performed for all patients after 36 hours of hCG stimulaton. rFSH dose >200 IU per one retrieved oocyte was considered as an indication for hyporesponse. Ten normo- and 9 hyporesponder patients were enrolled. Participants/materials, setting, methods Cells from the follicular fluid devoid of the cumulus oocyte complex were collected by centrifugation. RNA-seq was performed on the pooled cells of all participants. The study groups were compared by differential gene expression analysis by DESeq2. Single-cell RNA-seq was performed for >6000 individual cells of 3 normoresponder patients and cellular sub-populations were determined with Seurat package. CIBERTSORTx software was used to deconvolute the proportion of cell types from bulk RNA-seq data of all participants. Main results and the role of chance Alterations in the gene expression of ovarian somatic cells have been previously attributed to female age. We demonstrate that 407 genes are differentially expressed (FDR<0.05) between normo- and hyporesponder IVF patients after age adjustment. These genes were enriched into pathways of extracellular matrix reorganization (31 genes, FDR<0.005), post-translational protein phosphorylation (16 genes, FDR<0.006) and regulation of insulin-like growth factor transport and uptake by insulin-like growth factor binding proteins (16 genes, FDR<0.019). In addition, the sequencing of > 24 000 individual cells from 3 normoresponder patients revealed the somatic cell types of the preovulatory follicle in high resolution indicating the presence of immune cell populations (N = 4), epithelial cells, theca, cumulus cells and mural granulosa cell sub-populations (N = 7). Bioinformatic cell type deconvolution demonstrated significant differences between proportions of distinct cell sub-populations between normo- and hyporesponder patients and that several differentially expressed genes between the study groups can be attributed to a specific mural granulosa sub-population. Limitations, reasons for caution The number of study participants is small due to the high cost of the study methods. Wider implications of the findings The study demonstrates that hyporesponse to stimulation is associated with age-unrelated disturbances in gene expression of preovulatory somatic cells and that different somatic cell populations may have important underlying functions in the ovarian etiologies of infertility. Trial registration number not applicable