As a rising subject of interest in the field of exosomal contents with distant regulatory potency, exosomal circRNAs have received more and more attention in recent years [15]. Our study is the first to construct circRNA differential expression profiling in umbilical cord blood exosomes of PE patients as a starting point to explore the relationship between exosomal circRNAs and PE development. Based on the microarray data, we identified 304 differentially expressed circRNAs in umbilical cord blood exosomes of PE patients when compared with normal controls, including 143 up-regulated circRNAs and 161 down-regulated circRNAs, which suggested that the exosomal circRNA expression patterns in the PE samples were different from those in controls.
In subsequent validation experiments, the expression of exosomal circ_0077260 and circ_0090100 were significantly increased, and the expression of exosomal circ_0076206 were significantly decreased in PE samples. Growing evidence has shown that circRNAs can regulate parental gene expression through diverse mechanisms, such as transcription and splicing regulation, miRNA sponges, mRNA traps, translational modulation, and post-translational modification [22]. The parental genes of circ_0077260, circ_0090100 and circ_0076206 are CGA, SAT1 and MTCH1, respectively. Of these parental genes, CGA encodes for the common alpha subunit of four glycoprotein hormones, hCG (human chorionic gonadotropin), LH (luteinizing hormone), FSH (follicle-stimulating hormone) and TSH (thyroid-stimulating hormone) [23]. Previous studies have found that α-hCG is correlated with PE [24], and CGA was differentially expressed in placenta tissue among late-onset PE, early-onset PE and healthy controls [25, 26]. CGA is also considered as a novel estrogen receptor response gene in breast cancer and an outstanding candidate marker for predicting response to endocrine therapy [27]. Further studies are needed to determine whether the association among circ_0077260, CGA and estrogen is involved in the pathogenesis of PE.
The biological functions and potential pathways of these differential circRNAs were preliminarily predicted by GO and KEGG pathway analyses. Remarkably, several pathways were found to be significantly enriched, such as focal adhesion, glycosaminoglycan degradation, fatty acid metabolism, fatty acid biosynthesis and Notch signaling pathway. It is well known that focal adhesion is crucial to trigger cell adhesion and many other cellular processes including cell migration, spreading and proliferation [28], which are important in PE development. And localization studies in placental tissues have showed that cytotrophoblasts in all stages of differentiation express focal adhesion kinase [29]. In terms of metabolic process, PE has been demonstrated to be associated with increased insulin resistance, hypertriglyceridemia, high circulating free fatty acids, low high-density lipoprotein particles, and high maternal and fetal plasma amino acid concentrations [30]. These metabolic alterations may contribute to the pathophysiology of the syndrome and may also influence fetal growth. For Notch signaling pathway, defects in this pathway would have adverse effect on placentation. And it has been suggested that Notch pathway down-regulation is associated with PE [31]. Further constructed pathway network showed that the exchanges with these pathways largely depended on the existence of PI3K-Akt signaling pathway. The PI3K-Akt signaling pathway has been demonstrated to be a critical pathway mediating the growth-factor-dependent regulation of trophoblast growth and invasion [32]. The insufficient invasion of trophoblasts is known to be correlated with the development of PE [32]. Together, the altered circRNAs are associated with metabolic process, trophoblast growth and invasion related signaling pathways. Efficient biomarkers underlying these pathways need to be further investigated.
A large amount of evidence have indicated that exosomal circRNAs could act as ceRNA molecules or efficient miRNA sponges to regulate miRNA-targeted gene expression, transcription and protein synthesis [33–35]. The circRNAs may have many miRNA binding sites that competitively bind to miRNAs, and then alleviate the inhibitory effects of miRNAs on target molecules [21]. In our study, through circRNA/miRNA interactions analysis, we found that most of the exosomal circRNAs harbored miRNA binding sites, and some miRNAs were associated with PE. For example, miR-17-3p, miR-197, miR-424, miR-431 and miR-483 were reported to be aberrantly expressed in preeclamptic placenta [36–40]. miR-17-3p and miR-424-5p were matched with circ_0077260, which was verified to up-regulated in the umbilical cord blood exosomes of PE patients; miR-197-5p and miR-431-5p potentially binds to down-regulated circ_0076206; whereas miR-424-5p and miR-483-3p potentially matched with up-regulated circ_0090100. Specifically, exosomal miR-486-1-5p and miR-486-2-5p were reported to be up-regulated in PE pregnancy compared with normal pregnancy [41]. And miR-486-5p was matched with down-regulated circ_0076206. In addtion, miR-885-5p was increased in plasma from PE patients compared with healthy pregnant women, and it was released into circulation mainly inside exosomes [42], whereas miR-885-5p potentially matched with up-regulated circ_0077260. Therefore, we speculate that the role of exosomal circRNAs in PE development may be related to miRNA-mediated effects. The underlying mechanism of the circRNA-miRNA-target gene interaction in PE is worthy of further study.
In summary, our study firstly showed that exosomal circRNAs are aberrantly expressed in the umbilical cord blood of PE patients. Bioinformatics analysis further predicted the potential effects of these differentially expressed circRNAs and their interactions with miRNAs, highlighting the importance of exosomal circRNAs in the pathogenesis of PE and providing a basis for further studies on function and mechanism of exosomal circRNAs in PE development.