Circular RNA expression profiling of human granulosa cells during maternal aging reveals novel transcripts associated with assisted reproductive technology outcomes.

Circular RNAs (circRNAs) are a unique class of endogenous RNAs which could be used as potential diagnostic and prognostic biomarkers of many diseases. Our study aimed to investigate circRNA profiles in human granulosa cells (GCs) during maternal aging and to uncover age-related circRNA variations that potentially reflect decreased oocyte competence. CircRNAs in GCs from in vitro fertilization (IVF) patients with young age (YA, ≤ 30 years) and advanced age (AA, ≥ 38 years) were profiled by microarray, and validated in 20 paired samples. The correlation between circRNAs expression and clinical characteristics was analyzed in additional 80 samples. Chip-based analysis revealed 46 up-regulated and 11 down-regulated circRNAs in AA samples (fold change > 2.0). Specifically, circRNA_103829, circRNA_103827 and circRNA_104816 were validated to be up-regulated, while circRNA_101889 was down-regulated in AA samples. After adjustment for gonadotropin treatment, only circRNA_103827 and circRNA_104816 levels were positively associated with maternal age (partial r = 0.332, P = 0.045; partial r = 0.473, P = 0.003; respectively). Moreover, circRNA_103827 and circRNA_104816 expressions in GCs were negatively correlated with the number of top quality embryos (r = -0.235, P = 0.036; r = -0.221, P = 0.049; respectively). Receiver operating characteristic (ROC) curve analysis indicated that the performance of circRNA_103827 for live birth prediction reached 0.698 [0.570-0.825], with 77.2% sensitivity and 60.9% specificity (P = 0.006), and that of circRNA_104816 was 0.645 [0.507-0.783] (P = 0.043). Bioinformatics analysis revealed that both circRNAs were potentially involved in glucose metabolism, mitotic cell cycle, and ovarian steroidogenesis. Therefore, age-related up-regulation of circRNA_103827 and circRNA_104816 might be potential indicators of compromised follicular micro-environment which could be used to predict IVF prognosis, and improve female infertility management.