Circular RNA profile of parathyroid neoplasms: analysis of co-expression networks of circular RNAs and mRNAs

Abstract

Circular RNAs (circRNAs) are a recently identified class of non-coding RNAs that participate in multiple biological processes and tumour progression. However, circRNA expression pattern in parathyroid neoplasms remains unknown. The circRNA profile of 6 parathyroid carcinomas (PCs), 6 parathyroid adenomas (PAs) and 4 normal parathyroid tissues was assessed by a microarray. Bioinformatic analyses were performed to investigate potential core circRNAs via co-expression network. CircRNA and corresponding mRNA expression were validated in a cohort of parathyroid neoplasms by RT-qPCR and fluorescence in situ hybridization (FISH). Compared to normal parathyroid, 5310 and 1055 circRNAs were differentially expressed in PC and PA tissues, respectively. The differential expression of 4 circRNAs (hsa:circRNA_0035563 (p = 0.006), hsa:circRNA_0017545 (p = 0.009), hsa:circRNA_0001687 (p = 0.005) and hsa:circRNA_0075005 (p = 0.001)) and 4 mRNAs (MYC, FSCN1, ANXA2 and AKR1C3) between PC and PA tissues were confirmed by RT-qPCR. In addition, high expression of hsa:circ_0035563 was related to CDC73 mutations (p = 0.022) and recurrence in PC patients (p = 0.042). Furthermore, hsa:circ_0075005 helped distinguish PCs from benign lesions using FISH, and the area under the curve was 0.779 (p = 0.013). Our findings describe the circRNA profile of PC for the first time and suggest that circRNAs and mRNAs interact in parathyroid tumourigenesis. This study demonstrates that hsa:circ_0075005 and MYC mRNA may be used for the differential diagnosis of PC and PA. The expression levels of hsa:circ_0035563 are related to CDC73 mutations and recurrence in malignancy, highlighting the significance of this parameter in prognosis of PC patients.