Expression profile of circRNA in biliary atresia and choledochal cyst tissues : Chinese Medical Journal

Secondary Logo

Journal Logo


Expression profile of circRNA in biliary atresia and choledochal cyst tissues

Zhang, Wenjie; Wu, Zhouguang; Chen, Siqi; Zuo, Taoyan; Cheng, Zhen; Fu, Jingru; Wang, Bin

Editor(s): Guo, Lishao

Author Information
Chinese Medical Journal: October 12, 2022 - Volume - Issue - 10.1097/CM9.0000000000002079
doi: 10.1097/CM9.0000000000002079

To the Editor: Biliary atresia (BA) is a rare obliterative cholangiopathy that occurs during infancy in association with destructive liver inflammation and hepatic fibrosis. Although recent studies have shown that transforming growth factor (TGF-β) plays an important role in liver fibrosis and inflammation in BA, disease mechanisms remain poorly understood. Circular RNAs (circRNAs) are a recently discovered unique class of regulatory RNA molecules with high specificity. As previous evidence links circRNAs to the regulation of the TGF-β inflammatory signaling pathway, we further investigated the potential regulatory roles of circRNAs in BA and found novel insights into BA pathogenesis.

BA is characterized by destructive inflammation and fibrosis in the affected liver, destroying the extrahepatic bile ducts and disrupting bile flow. If not treated in time, BA can rapidly worsen and result in devastating consequences. Surgical intervention, known as Kasai hepatoportoenterostomy, is the most common treatment option for patients with BA. Unfortunately, in 40% to 50% of patients, bile drainage fails to improve following surgery, and the majority eventually requires liver transplantation. As a key regulator and indicator of the inflammatory response, TGF-β is upregulated in infants with BA.[1] Hence, understanding TGF dysregulation could provide a better understanding of the disease mechanism and consequently provide new insights for treatment development.

CircRNAs are a unique class of noncoding RNA that lack 5′-3′ structure and polyA tails and showed regulatory roles of circRNAs in the TGF-β signaling pathway.[2] As the hyperactivation of TGF-β signaling pathways has been shown to be associated with the activation of hepatic stellate cells and ultimately the regulation of hepatic fibrosis, circRNAs may be involved in the dysregulation of TGF-β signaling during BA. Therefore, investigating the circRNA-miRNA-TGF-β pathway could reveal the pathogenesis of BA.

In this study, circRNA sequencing was performed in both BA and choledochal cyst (CC) tissue samples. An altered expression of these circRNAs was confirmed using reverse transcription quantitative polymerase chain reaction (RT-qPCR), followed by receiver operating characteristic (ROC) analysis using clinical data to evaluate the diagnostic and prognostic potentials of the circRNAs of interest.

A total of 109 patient samples obtained between January 2019 and December 2020 were included in the present study. The selected patients underwent Kasai hepatoportoenterostomy at Shenzhen Children's Hospital. Liver tissues from 68 patients with BA and 41 patients with CC were immediately stored in RNA later (ThermoFisher, AM7020, Waltham, MA, USA) at −80°C. Sample collection was performed after receiving the consent of patients and relatives, with approval from the ethics committee of Shenzhen Children's Hospital (No. SUMC2019-119).

Total RNA was extracted using TRIzol (MRC, TR118-500, Cincinnati, OH, USA) according to the manufacturer's instructions. Then, RNA was sequenced as per a previous study.[3] A total of 2 μg RNA was reverse-transcribed into cDNA using M-MLV reverse transcriptase (Promega, Madison, WI, USA). The resulting cDNAs were used for qPCR with primers shown in Supplementary Table 1, qPCR was performed using ChamQ SYBR qPCR Master Mix (Vazyme, Nanjing, China) according to the manufacturer's instructions.

To analyze the Illumina HiSeq data, Find_circ software (version 1.2, Max-Delbrück-Center for Molecular Medicine, Berlin, Germany) was used to compare the sequencing data and reference genes, followed by a comparison with circBase for known circRNAs. The normalized expression levels were analyzed using the DEGseq package in R (version 3.6.3, Tsinghua University, Beijing, China). Expression levels were considered different among two groups if P < 0.05 (corrected by Benjamini and Hochberg algorithm) and |log2FC|≥1. Graphs were generated using the ggplot2 package in R. The putative circRNA-miRNAs-TGF-β pathway interaction networks were constructed representatively using miRanda3.3a (Memorial Sloan Kettering Cancer Center, NY, USA) and Cytoscape3.5.1 (National Institute of General Medical Sciences, Washington, USA). The Mann-Whitney U test was used to compare differences in circRNA expression in RT-qPCR between the two groups using GraphPad Prism 7.0 software (La Jolla, CA, USA). P < 0.05 was considered significant. ROC analysis was used to analyze the specificity and sensitivity of hsa_circ_0009096 for BA.

Three samples each of BA and CC tissues were used for circRNA sequencing. CircRNAs documented in circBase ( with more than a two-fold difference in expression level were further analyzed [Figure 1A]. Under these criteria, 90 differentially expressed circRNAs were identified. In particular, in the BA group, 50 circRNAs were upregulated whereas 40 were downregulated, relative to the CC group. Then, we constructed an interaction network using known competitive endogenous RNA (ceRNA) mechanisms to predict interactions between circRNAs and miRNAs in the TGF-β signaling pathway [Figure 1B]. In addition, differentially expressed circRNAs with a length of 500 to 3000 bps and multiple binding sites were selected for further analysis, including six circRNAs, for example, four upregulated circRNAs, namely, hsa_circ_0009096, hsa_circ_0006961, hsa_circ_0021412, and hsa_circ_0001348; and two downregulated circRNAs, namely, hsa_circ_0072697 and hsa_-circ_0025498. As shown in Figure 1C and Supplementary Figure 1,, all six circRNAs were circular and expressed in liver tissues.

Figure 1:
circRNA sequencing and validation of BA and CC tissues. (A) Heatmap of differentially expressed circRNAs between BA and CC tissues. Selected circRNAs showed more than a two-fold difference in expression levels, had a P value of < 0.05 and were documented in the circBase database. (B) ceRNA network of circRNA-miRNA-TGF-β signaling. (C) Circularity confirmation of hsa_circ_0009096 using RNA electrophoresis and Sanger sequencing. (D) RT-qPCR with a larger sample group showed higher expression levels of hsa_circ_0009096 in the BA group (n = 66) than in the CC group (n = 38). P < 0.01. (E) ROC analysis indicating the predictive value of hsa_circ_0009096 in BA diagnosis with high sensitivity and high specificity, AUC = 0.8008, P < 0.0001. BA: Biliary atresia; CC: Choledochal cyst; circRNAs: Circular RNAs; ceRNA: competitive endogenous RNA; miRNA: microRNA; ROC: Receiver operating characteristic; RT-qPCR: Reverse transcription quantitative polymerase chain reaction; TGF-β: Transforming growth factor-β.

Moreover, both hsa_circ_0009096 and hsa_circ_0072697 showed higher expression levels in the BA group; the expression level of hsa_circ_0009096 remained consistent with our circRNA sequencing analysis, whereas hsa_-circ_0072697 was downregulated in BA samples [Supplementary Figure 2,]. To further confirm the high expression of hsa_circ_0009096, an additional RT-qPCR was performed using a larger sample group, including 65 patients with BA and 38 patients with CCs, without any overlap with circRNA sequencing. The hsa_circ_0009096 level was significantly higher in the BA group than in the CC group for the additional RT-qPCR experiment with larger sample sizes [Figure 1D]. Furthermore, ROC analysis showed that hsa_circ_0009096 has good specificity and sensitivity for diagnosing BA [Figure 1E]. The characteristics of patients with BA or CC before Kasai portoenterostomy are shown in Supplementary Table 2,

In the present study, circRNA sequencing showed differential expression levels in 90 circRNAs in the BA and CC groups. Six of these circRNAs have predicted regulatory roles in the TGF-β signaling pathway, potentially contributing to inflammatory responses. In particular, hsa_-circ_0009096 showed predictive values for BA diagnosis. Our study identified circRNAs with potential regulatory roles in the TGF-β signaling pathway to further elucidate BA pathogenesis. Unfortunately, not much is known about the functional role of hsa_circ_0009096. Future studies are needed to understand the mechanism of hsa_circ_0009096.


This research was supported by Sanming Project of Medicine in Shenzhen under [grant number SZSM201812055]; National Natural Science Foundation of China under [grant number 81770512]; Research topic on academic and postgraduate education in China under [grant number 2020MSA126].


1. Ortiz-Perez A, Donnelly B, Temple H, Tiao G, Bansal R, Mohanty SK. Innate immunity and pathogenesis of biliary atresia. Front Immunol 2020;11:329–342. doi: 10.3389/fimmu.2020.00329.
2. Hu W, Han Q, Zhao L, Wang L. Circular RNA circRNA_15698 aggravates the extracellular matrix of diabetic nephropathy mesangial cells via miR-185/TGF-(1. J Cell Physiol 2019;234:1469–1476. doi: 10.1002/jcp.26959.
3. Zheng ML, Du XP, Zhao L, Yang XC. Expression profile of circular RNAs in epicardial adipose tissue in heart failure. Chin Med J 2020;133:2565–2572. doi: 10.1097/cm9.0000000000001056.

Supplemental Digital Content

Copyright © 2022 The Chinese Medical Association, produced by Wolters Kluwer, Inc. under the CC-BY-NC-ND license.