Deconvolution at the single-cell level reveals ovarian cell-type-specific transcriptomic changes in PCOS

Abstract

Abstract Background Polycystic ovary syndrome (PCOS) is one of the most common reproductive endocrine disorders in females of childbearing age. Various types of ovarian cells work together to maintain normal reproductive function, whose discordance often takes part in the development and progression of PCOS. Understanding the cellular heterogeneity and compositions of ovarian cells would provide insight into PCOS pathogenesis, but are, however, not well understood. Transcriptomic characterization of cells isolated from PCOS cases have been assessed using bulk RNA-seq but cells isolated contain a mixture of many ovarian cell types. Methods Here we utilized the reference scRNA-seq data from human adult ovaries to deconvolute and estimate cell proportions and dysfunction of ovarian cells in PCOS, by integrating various granulosa cells(GCs) transcriptomic data. Results We successfully defined 22 distinct cell clusters of human ovarian cells. Then after transcriptome integration, we obtained a gene expression matrix with 13,904 genes within 30 samples (15 control vs. 15 PCOS). Subsequent deconvolution analysis revealed decreased proportion of small antral GCs and increased proportion of KRT8 high mural GCs, HTRA1 high cumulus cells in PCOS, especially increased differentiation from small antral GCs to KRT8 high mural GCs. For theca cells, the abundance of internal theca cells (TCs) and external TCs was both increased. Less TCF21 high stroma cells (SCs) and more STAR high SCs were observed. The proportions of NK cells and monocytes were decreased, and T cells occupied more in PCOS and communicated stronger with inTCs and exTCs. In the end, we predicted the candidate drugs which could be used to correct the proportion of ovarian cells in patients with PCOS. Conclusions Taken together, this study provides insights into the molecular alterations and cellular compositions in PCOS ovarian tissue. The findings might contribute to our understanding of PCOS pathophysiology and offer resource for PCOS basic research.

Keywords

• Polycystic ovary syndrome
• Deconvolution
• Granulosa cells
• scRNA-seq
• Bulk RNA-seq