Exosomal RNA Expression Profiles and Their Prediction Performance in Patients With Gestational Diabetes Mellitus and Macrosomia

INTRODUCTION: Exosomes are cell-derived vesicles that are present in many biological fluids. Exosomal RNAs in cord blood may allow intercellular communication between mother and fetus. We aimed to establish exosomal RNA expression profiles in cord blood from patients with gestational diabetes mellitus and macrosomia (GDM-M) and evaluate their prediction performance. METHODS: We used microarray technology to establish the differential messenger RNA (mRNA), long non-coding RNA (lncRNA), and circular RNA (circRNA) expression profiles in cord blood exosomes from 3 patients with GDM-M compared with 3 patients with GDM and normal neonatal weight, followed by qPCR validation in an additional 40 patients with GDM. Logistic regression, receiver operating characteristic (ROC) curves, and graphical nomogram were applied to evaluate the performance of exosomal RNA (in peripheral blood) in macrosomia prediction. RESULTS: A total of 98 mRNAs, 372 lncRNAs, and 452 circRNAs were differentially expressed in cord blood exosomes from patients with GDM-M. Pathway analysis based on screening data showed that the differential genes were associated with Phosphatidylinositol 3'-kinase (PI3acK)-Akt signaling pathway, Janus kinase/signal transducers and activators of transcription (JAK/STAT) signaling pathway, Transforming growth factor (TGF)-beta signaling pathway, insulin resistance, glycerolipid metabolism, fatty acid degradation, and mammalian target of rapamycin (mTOR) signaling pathway. After validation by qPCR, the expressions of GDF3, PROM1, AC006064.4, lnc-HPS6-1:1, and circ_0014635 were significantly increased and the expression of lnc-ZFHX3-7:1 was significantly decreased in cord blood exosomes of an additional 20 patients with GDM-M. The risk prediction performance of the expression of these validated genes (in peripheral blood exosomes) for GDM-related macrosomia was also evaluated. Only GDF3 expression and AC006064.4 expression showed well prediction performance [area under the curve (AUC) = 0.78 and 0.74, respectively]. Excitingly, the model including maternal age, fasting plasma glucose, 2-h plasma glucose, GDF3 expression, and AC006064.4 expression in peripheral blood exosomes had better prediction performance with an AUC of 0.86 (95% CI = 0.75–0.97). CONCLUSION: These results showed that exosomal RNAs are aberrantly expressed in the cord blood of patients with GDM-M and highlighted the importance of exosomal RNAs in peripheral blood for GDM-M prediction.