Transcriptome Analysis of Mesenchymal Progenitor Cells Revealed Molecular Insights into Metabolic Dysfunction and Inflammation in Polycystic Ovary Syndrome

Abstract

Polycystic ovary syndrome (PCOS) is a female endocrine disorder with metabolic issues. Hyperandrogenism combined with hyperinsulinemia exacerbates the reproductive, metabolic, and inflammatory problems in PCOS patients. The etiology of PCOS is unclear. Patient-specific induced pluripotent stem cells (iPSCs) offer a promising model for studying disease mechanisms and conducting drug screening. Here, we aim to use mesenchymal progenitor cells (MPCs) derived from PCOS iPSCs to explore the mechanism of PCOS. We compared the transcriptome profiles of PCOS and healthy control (HC) iPSC-derived MPCs (iPSCMs). Moreover, we assess the impact of androgens on iPSCMs. In the comparison between PCOS and HC, the expression levels of 1026 genes were significantly different. A gene set enrichment analysis (GSEA) revealed that adipogenesis- and metabolism-related genes were downregulated, whereas inflammation-related genes were upregulated in the PCOS iPSCMs. Dysregulation of the TGF-β1 and Wnt signaling pathways was observed in the PCOS iPSCMs. Furthermore, there was impaired adipogenesis and decreased lipolysis in the PCOS iPSCMs-derived adipocytes. With testosterone treatment, genes related to metabolism were upregulated in the HC iPSCMs but downregulated in the PCOS iPSCMs. The impact of testosterone varied among HCs and PCOS iPSCMs, possibly because of a genetic predisposition toward PCOS. This study found specific signaling pathways that could serve as therapeutic targets for PCOS.

Keywords: RNA sequencing; induced pluripotent stem cells; polycystic ovary syndrome.

Figure 1

Characterization of induced pluripotent stem cells (iPSCs) and iPSC-derived mesenchymal progenitor cells (iPSCMs). (a) Alkaline phosphatase staining showing alkaline phosphatase activity in iPSCs (blue). (b) Immunostaining for the pluripotency markers NANOG (red) and TRA-1-60 (green) in iPSCs. NANOG was expressed in the nuclei, while TRA-1-60 was expressed on the cell surface. Nuclei were stained with Hoechst 33,342 (blue). (c) Hematoxylin and eosin (H&E) staining of teratomas derived from iPSCs. The markers show the sites of lineage-specific morphological characteristics, including neuroepithelial rosette (NR) for ectoderm, chondrocyte (C) for mesoderm, and intestinal or gut-like epithelium (I/E) for endoderm. (d) Bright-field image showing the fibroblast-like spindle morphology of iPSCMs. (e) Oil red O staining of lipid droplets in adipocytes derived from iPSCMs. (f) Alcian blue staining for acidic polysaccharides in chondrocytes derived from iPSCMs. (g) Alizarin Red S staining of calcium nodules in osteoblasts derived from iPSCMs. Scale bars, 100 mm.

Figure 2

Transcriptomic analyses of PCOS and HC iPSCMs. (a) The volcano plot depicts the results of the differential expression analysis comparing PCOS and HC iPSCMs, highlighting the differentially expressed genes (DEGs), with red indicating upregulated genes and blue indicating downregulated genes. (b) Principal component analysis (PCA) using DEGs illustrating distinct expression profiles between PCOS and HC iPSCMs. (c) Expression of DEGs related to the insulin response across all samples. (d) The bar plot displays the significantly enriched hallmark gene sets. All gene sets with p < 0.05 were derived from a gene set enrichment analysis (GSEA). HC: healthy control.

Figure 3

Transcriptomic analysis revealed adipogenesis dysfunction in PCOS iPSCMs. (a–d) GSEA plots of gene sets related to adipogenesis-related pathways (a), TGF-β1-related pathways (b), BMP- and Wnt-related pathways (c), and C/EBPA-related pathways (d). p < 0.05 for all gene sets. (e) Expression of genes related to adipogenesis across all samples. (f) Oil red O staining showing iPSCM-derived adipocytes with lipids. Scale bars, 100 μm. (g) Flow cytometry quantification of iPSCM-derived adipocytes with lipid droplets. The experiments were conducted in biological triplicate. Error bar represents the mean ± SD. t-test: * p < 0.05. HC: healthy control.

Figure 4

The suppression of fatty acid and glucose metabolism pathways in PCOS iPSCMs. (a–c) GSEA plots reveal the expression of gene sets associated with fatty acid metabolism (a), glucose metabolism (b), and the cellular response to insulin stimulation pathways (c) in iPSCMs from individuals with PCOS compared to those from the HCs. p < 0.05 for all gene sets. (d) Heatmap illustrating the expression of genes related to metabolic pathways across all samples. (e) Isoproterenol-induced glycerol release normalized to triglyceride levels in iPSCM-derived adipocytes. The experiments were conducted in biological triplicate. Error bar represents the mean ± SD. t-test: ** p < 0.01. TG: triglyceride, VEH: vehicle, ISO: isoproterenol, HC: healthy control.

Figure 5

The increase in the expression of immune-related genes in PCOS iPSCMs. (a–c) GSEA plots illustrating the expression of gene sets associated with TNF and NFκB signaling (a), interleukin and interferon responses (b), and cytokine and chemotaxis elements between iPSCMs from PCOS patients and HCs (c). p < 0.05 for all gene sets. (d) Relative expression of cytokines across all samples. (e) ICAM1, VEGFA, and VCAM1 expression in the granulosa cells were analyzed via real-time quantitative PCR. One control sample was used for normalization. The experiments were conducted in biological triplicate. Error bar represents the mean ± SD. t-test: * p < 0.05, ** p < 0.01. HC: healthy control.

Figure 6

Transcriptomic analysis revealing the impact of testosterone on iPSCMs from PCOS patients and HCs. (a) Venn diagram depicting differentially expressed genes (DEGs) in PCOS and HC iPSCMs with and without testosterone treatment. (b,c) The bar plot displays the significantly enriched hallmark gene sets in HC iPSCMs with vs. without testosterone treatment (b) and in PCOS iPSCMs with vs. without testosterone treatment (c). All gene sets with p < 0.05 were derived from the GSEA. (d,e) Gene set variation analysis (GSVA) showing the activity of hallmark gene sets related to adipogenesis (d), fatty acid metabolism (e), and the inflammatory response (f) for each sample. Green: cells without testosterone treatment, jasmine: cells with testosterone treatment, squares: healthy control, circles: PCOS patient. The p values were derived from a paired t-test. HC_DN: genes whose expression was downregulated in HC iPSCMs with testosterone treatment vs. those without testosterone treatment, HC_UP: genes whose expression was upregulated in HC iPSCMs treated with testosterone vs. those without testosterone, PCOS_DN: genes whose expression was downregulated in PCOS iPSCMs with vs. without testosterone treatment, PCOS_UP: genes whose expression was upregulated in PCOS iPSCMs with vs. without testosterone treatment, HC: healthy control, w/o: without, w/: with.

Figure 7

The causes of inflammation in PCOS. Both hyperandrogenism and adipocyte hypertrophy contribute to inflammation in patients with PCOS. However, inflammation can occur due to dysregulation in inflammatory genes independent of hyperandrogenism and adipocyte hypertrophy. Dysregulation in BMPs, TGF-β, and Wnt may be involved in adipocyte hypertrophy and inflammation in polycystic ovary syndrome (PCOS), as observed in PCOS iPSCMs.