N6-methyladenosine regulator-mediated methylation modification patterns and immune infiltration characterization in Polycystic Ovary Syndrome (PCOS)

• Main Authors: Sihan Zhou (Author), Rui Hua (Author), Song Quan (Author)
• Format: Book
• Published: BMC, 2023-04-01T00:00:00Z.
• Subjects: article
• Online Access: Connect to this object online.
• Notes: 10.1186/s13048-023-01147-9; 1757-2215
• Keywords: Polycystic ovary syndrome; N6-methyladenosine; M6A regulators; Bioinformatics analysis; Gynecology and obstetrics; RG1-991
• Source: Journal of Ovarian Research, Vol 16, Iss 1, Pp 1-15 (2023)

Abstract Background Polycystic ovary syndrome (PCOS) is a multisystem-related disease whose pathophysiology is still unclear. Several regulators of N6-methyladenosine (m6A) modification were confirmed to play a regulatory role in PCOS. Nonetheless, the roles of m6A regulators in PCOS are not fully demonstrated. Materials and methods Four mRNA expression profiling microarrays were obtained from the Gene Expression Omnibus (GEO) database. Differentially expressed m6A regulators between PCOS and normal patients were identified by R software. A random forest modal and nomogram were developed to assess the relationship between m6A regulators and the occurrence risk of PCOS. A consensus clustering method was utilized to distinctly divide PCOS patients into two m6A subtypes (m6A cluster A/B). The patterns of differential expression and immune infiltration were explored between the two m6A clusters. Results In this study, 22 significant m6A regulators were identified between healthy controls and PCOS patients. The random forest model determined three optimal m6A regulators which are related to the occurrence risk of PCOS, including YTHDF1, RBM15 and METTL14. A nomogram was established based on these genes, and its predictive reliability was validated by decision curve analysis. The consensus clustering algorithm distinctly divided PCOS cases into two m6A subtypes. The ssGSEA algorithm found that the immune infiltration was markedly enriched in m6A cluster B than in cluster A. The m6A-pattern related differentially expressed genes (DEGs) of the two m6A subtypes were demonstrated by differential expression analysis. We found that they were enriched in immune-related genes and various infection pathways. Based on the m6A-pattern related DEGs, the PCOS patients were classified into two m6A-pattern related genomic subtypes (gene clusters A and B). Conclusions The present study provided evidence concerning the different modification patterns of m6A regulators in PCOS compared with normal patients. This study will help clarify the overall impact of m6A modification patterns and related immune infiltration on PCOS.