Intrauterine Reprogramming of the Polycystic Ovary Syndrome: Evidence from a Pilot Study of Cord Blood Global Methylation Analysis

Abstract

Polycystic ovary syndrome (PCOS) affects 5-15% of women. PCOS is a heterogeneous disorder displaying endocrine, metabolic, and reproductive dysfunction and cardiovascular risk manifestations. Evidence of heritability exists, but only a portion of the genetic transmission has been identified by genome-wide association studies and linkage studies, suggesting epigenetic phenomena may play a role. Evidence implicates intrauterine influences in the genesis of PCOS. This was a pilot study that aimed at identifying an epigenetic PCOS reprogramming signature by profiling the methylation of the DNA extracted from umbilical cord blood (UCB) from 12 subjects undergoing in vitro fertilization. Six subjects were anovulatory PCOS women diagnosed by Rotterdam criteria and six ovulatory non-PCOS women matched for age and body mass index. UCB was collected at delivery of the placenta; the DNA was extracted and submitted to methylation analysis. A differential methylation picture of prevalent hypomethylation affecting 918 genes was detected. Of these, 595 genes (64.8%) carried single or multiple hypomethylated CpG dinucleotides and 323 genes (35.2%) single or multiple hypermethylated CpG dinucleotides. The Ingenuity Pathway Analysis (IPA) online platform enlisted 908 of the 918 input genes and clustered 794 of them into 21 gene networks. Key features of the primary networks scored by IPA included carbohydrate and lipid metabolism, neurotransmitter signaling, cardiovascular system development and function, glycosaminoglycan signaling regulation and control of amino acid biosynthesis. Central to the network activities were genes controlling hormonal regulation (ESR1), mitochondrial activity (APP, PARK2), and glucose metabolism (INS). Regulatory pathways such as G-protein coupled receptor signaling, inositol metabolism, and inflammatory response were also highlighted. These data suggested the existence of a putative "PCOS epigenomic superpathway" with three main components: glucotoxic, lipotoxic, and inflammatory. If our results are confirmed, they hint at an epigenetic at risk PCOS "signature" may thus exist that may be identifiable at birth. Additional studies are needed to confirm the results of this pilot study.

Keywords: diabetes; epigenetics; metabolic syndrome; polycystic ovary syndrome; pregnancy.

Figure 1

Distribution of the differentially methylated CpG dinucleotides in the umbilical cord blood from offspring of polycystic ovary syndrome mothers vs controls with a minimum twofold methylation change. Top panel all differentially methylated CpG dinucleotides; mid panel hypermethylated CpG dinucleotides; bottom panel hypomethylated CpG dinucleotides. The exploded portion of the pie charts reports the classification of those differentially methylated CpG dinucleotides that map in or near know genes. The other portion of the pie chart reports the main characteristics of the differentially methylated CpG dinucleotides that do not map in or near known genes.

Figure 2

Ingenuity pathway analysis (IPA) network and canonical pathway analysis for the genes included in network (A). Depiction of network 1. The main nodes are identified with bigger bolded font (see Table 2 for details). Gene/domain symbol shading: green, hypomethylated; red, hypermethylated; white, no differential methylation. For hyper- and hypomethylated genes/domains, red/green gradient relates to the methylation level. (B) Depiction of the two main canonical pathway webs of network 1 highlighted by IPA. The first (left) appears principally devoted to the regulation of the trafficking of hormones, proteins, cholesterol, fatty acid, and glucose with an important hormonal involvement. The second (right) is instead centered on the lipid metabolism and mitochondrial functioning regulation. Canonical pathway red symbol gradient relates to the p-value for the likelihood of the association between the differentially methylated genes in our experiment and the pathway. The smaller the p-value, the darker the red shading and the stronger the association. For a detailed explanation of molecule shapes and relationship types, see http://qiagen.force.com/KnowledgeBase/articles/Basic_Technical_Q_A/Legend.

Figure 3

Polycystic ovary syndrome (PCOS) superpathway. (A) Full superpathway. (B) Glucotoxic PCOS component; (C) Lipotoxic PCOS component; (D) Inflammatory PCOS component. The main genes/domain common to the three components are identified with bigger bolded font. The design of the PCOS superpathway has been conducted in ingenuity pathway analysis by linking statistically significant key genes and canonical pathway highlighted by networks 1–10. Gene/domain symbol shading: green, hypomethylated; red, hypermethylated; white, no differential methylation. For hyper- and hypomethylated genes/domains, red/green gradient relates to the methylation level.