PMID：31148183.pdf

Accepted ArticleThis article has been accepted for publication and undergone full peer review but has not been through the copyediting,typesetting,pagination and proofreading process,which may lead to differences between this version and the Version of Record.Please cite this article as doi:10.1002/hep.30795 This article is protected by copyright.All rights reserved.Article type :Original A Noncoding Regulatory RNAs Network Driven by Circ-CDYL Acts Specifically in the Early Stages Hepatocellular Carcinoma Yanping Wei1,2,3,10,Xin Chen1,2,10,Chi Liang1,2,10,Yan Ling1,2,10,Xinwei Yang4,Xiaofei Ye5,Hailing Zhang6,Pinghua Yang4,Xiuliang Cui1,2,Yibing Ren1,2,Xianglei Xin4,Hengyu Li6;Ruoyu Wang4;Wenjing Wang7,Feng Jiang8,Suiyi Liu4,Jing Ding1,2,Baohua Zhang4,Liang Li1,2*and Hongyang Wang 1,2,9*1.International Co-operation Laboratory on Signal Transduction,Eastern Hepatobiliary Surgery Institute,Second Military Medical University,200438,Shanghai,China.2.National Center for Liver Cancer,Shanghai,China.3.The Graduate School of Fujian Medical University.4.Eastern Hepatobiliary Surgery Hospital,Second Military Medical University,200438,Shanghai,China.5.Department of Health Statistics,Second Military Medical University.6.Changhai Hospital,Second Military Medical University,200438,Shanghai,China.7.Shanghai Municipal Center for Disease Control&Prevention.8.Jiangsu Key Laboratory of Molecular and Translational Cancer Research,Cancer Hospital of Jiangsu Province,Nanjing,China.9.National Laboratory for Oncogenes and Related Genes,Cancer Institute,RenJi Hospital,Shanghai Jiao Tong University,200441,Shanghai,China.10.Co-first author Accepted ArticleThis article is protected by copyright.All rights reserved.*Correspondence:Hongyang Wang()and Liang Li()are correspondence authors.KEYWORDS HCC;Circular RNA;Early diagnosis;Early treatment;MicroRNAs sponge.Address Correspondence to lead contact:Hongyang Wang,M.D.Academician of Chinese Academy of Engineering Professor&Director National Center for Liver Cancer,China International Cooperation Laboratory on Signal Transduction Eastern Hepatobiliary Surgery Hospital,Shanghai Vice President of Chinesse Anti-Cancer Association 225 Changhai Road,Shanghai 200438,China E-mail:,Tel:86 21 8187 5361,Fax:86 21 6556 6851.Liang Li,Ph.D.Associate Professor,International Co-operation Laboratory on Signal Transduction,Eastern Hepatobiliary Surgery Institute,Second Military Medical University,225 Changhai Road,200438 Shanghai,China;Accepted ArticleThis article is protected by copyright.All rights reserved.National Center for Liver Cancer,Shanghai,China;E-mail:,Tel:86 21 8187 5364.ABBREVIATIONS AUC:area under the curve;AKT:AKT serine/threonine kinase;BIRC5:Baculoviral IAP repeat containing 5;CeRNA:competing endogenous RNA;CircRNA:circular RNA;CDYL:chromodomain Y like;CTNNB1:catenin beta 1;Co-IP:Co-immunoprecipitation;EPCAM:epithelial cell adhesion molecule;HCC:hepatocellular carcinoma;HDGF:hepatoma-derived growth factor;HIF1AN:hypoxia inducible factor asparagine hydroxylase;HEY-1:hes related family bHLH transcription factor with YRPW motif 1;ICD:intracellular domain;mTOR:mechanistic target of rapamycin kinase;microRNAs:microRNAs;MREs:microRNAs response elements;NCL:nucleolin;NOD:N-Oxalyl-D-alanine;OR:odds ratio;PI3K:phosphatidylinositol-4,5-bisphosphate 3-kinase;ROC:receiver operating characteristics;T-ICs:tumor-initiating cells AUTHOR CONTRIBUTIONS Hongyang Wang contributed to study concept design and supervised all works;Liang Li contributed to analysis and interpretation of data and drafting of the manuscript.Yanping Wei,Xin Chen,Chi Liang and Yan Ling contributed to acquisition of data.Xinwei Yang,Xiaofei Ye,Hailing Zhang,Xiuliang Cui,Yibing Ren,Pinghua Yang,Xianglei Xin,Hengyu Li;Accepted ArticleThis article is protected by copyright.All rights reserved.Ruoyu Wang;Wenjing Wang,Feng Jiang,Suiyi Liu,Jing Ding and Baohua Zhang contributed to technical and material support.FUNDING We gratefully acknowledge the support from the State Key Project on Infectious Diseases of China(2018ZX10723204-002-002),National Natural Science Foundation of China(81372674,81830054 and 81672777),Key Research and Development Projects of China(2016YFC1101402),Priority Among Priorities Clinical Medical Center Project of Shanghai(2017ZZ01007),Shanghai Rising-Star Program(17QA1405700)and Clinical Technology Innovation Project of Shanghai Shenkang Hospital Development Center(SHDC12016127).STATEMENT No conflicts of interest exist.The authors have no financial relationship to disclose.Word count:5990 words including abstract,introduction,experimental procedure,results,discussion and references.Accepted ArticleThis article is protected by copyright.All rights reserved.ABSTRACT Hepatocellular carcinoma(HCC)is the fastest-rising cause of cancer-related death worldwide,but its deficiency of specific biomarkers and therapeutic targets in the early stages lead to severe inadequacy in the early diagnosis and treatment of HCC.Covalently closed circular RNA,which was once considered an aberrant splicing byproduct,is now drawing new interest in cancer research due to its remarkable functionality.Beneath the surface of the dominant functional proteins events,a hidden circRNA-centric noncoding regulatory RNAs network active in the very early stage of HCC is here revealed by a genome-wide analysis of mRNA,circRNA and microRNA expression profiles.Circ-CDYL is specifically upregulated in the early stages of HCC and therefore contributes to the properties of EPCAM-positive liver tumor-initiating cells.Circ-CDYL interacts with mRNAs encoding hepatoma-derived growth factor(HDGF)and hypoxia inducible factor asparagine hydroxylase(HIF1AN)by acts as the sponge of miR-892a and miR-328-3p,respectively.Subsequently,activation of the PI3K-AKT-mTORC1/-catenin and NOTCH2 pathways,which promote the expression of the effect proteins baculoviral IAP repeat containing 5(BIRC5 or SURVIVIN)and MYC proto-oncogene,is influenced by Circ-CDYL.A treatment incorporating Circ-CDYL interference and traditional enzyme inhibitors targeting PI3K and HIF1AN demonstrated highly effective inhibition of stem-like characteristics and tumor growth in HCC.Finally,we demonstrated that the Circ-CDYL expression or which combined with HDGF and HIF1AN are both independent marker for discrimination of early stages HCC with the odds ratio(OR)of 1.09(95%CI:1.02-1.17)and 124.58(95%CI:13.26-1170.56),respectively.Conclusion.Accepted ArticleThis article is protected by copyright.All rights reserved.These findings uncover a circRNA-centric noncoding regulatory RNAs network in the early stages HCC and thus provide a possibility for surveillance and treatment of HCC early.INTRODUCTION Liver cancer is ranked as the second leading cause of cancer-related mortality worldwide in man(1).Despite the success of treatments for some other forms of cancer,the incidence rate and death rate of liver cancer are both increasing rapidly in the United States(2).Accumulating data have revealed that the survival rates of hepatocellular carcinoma(HCC)patients vary with differences in neoplasm staging.The 5-year survival rate for HCC patients is disappointingly low at 14%overall(3).In contrast,very early-stage(Barcelona Clinic Liver Cancer stage 0,BCLC stage 0)and early-stage(BCLC stage A)HCC patients have 5-year survival rates of 90%and 50%-70%after surgical treatment,respectively(4-6).Subgroup analysis has shown that BCLC stage A patients have an optimal median survival of 53 months,while patients at BCLC stages B,C,and D have median survival times of only 16,7,and 3 months,respectively(7).Unfortunately,the lack of specific symptoms and accurate molecular targets for HCC in early stages leads to serious obstacles in the early diagnosis and treatment of HCC(8).Two-thirds of HCC patients are first diagnosed at an advanced stage,at which little time is left to perform any effective therapy(9).Considering that the genomic profile differs in each cancer stage(10),and the present literature has mainly focused on the late stages of HCC,it is urgent to study the early stages of HCC directly.Accepted ArticleThis article is protected by copyright.All rights reserved.Approximately 70%-90%of the human genome is transcribed into RNAs,but surprisingly,less than 2%of the total genome is protein-coding genes(11).The vast amount of noncoding RNA was long considered“transcriptional noise”with little functional potential.More recently,accumulating studies have demonstrated that these noncoding RNAs,including microRNAs,long noncoding RNAs(lncRNAs)and the recently rediscovered circular RNAs(circRNAs),have broad effects on physiological and pathological processes in humans(11)through a previously unknown mechanism called competing endogenous RNA(ceRNA)interaction or microRNA sponge regulation.This hypothesis demonstrates that messenger RNAs,pseudogenes and noncoding RNAs perform“cross talk”via common microRNA response elements(MREs)(12).Covalently closed circRNAs were discovered nearly 40 years ago(13)and were often considered aberrant splicing byproducts with little functional potential.Until recently,the emerging roles of circRNAs in physiological and pathological conditions were characterized(14).CircRNAs,which are unusually stable noncoding RNA molecules,are considered a result of proactive back-splice events,in which the 3 tail of one exon is joined to the 5 head of an upstream exon(15).The known functions of circRNAs include sponging microRNAs or proteins,modulation of transcription and splicing,and translation as polypeptides(15).However,the molecular and biological functions of most circRNAs remain unexplored.Accepted ArticleThis article is protected by copyright.All rights reserved.Here,by taking the very early-stage HCC as a direct research object,we constructed a circRNA-centric noncoding regulatory RNAs network through a genome-wide analysis of the mRNA,circRNA and microRNA expression profiles of BCLC stage 0 HCC.Furthermore,we discuss potential early diagnostic and therapeutic implications for HCC by exploring the relevant functions and underlying mechanism of the key noncoding regulatory RNAs network.MATERIALS AND METHODS HCC patients and clinical samples HCC tissues and matched non-tumorous adjacent tissues were obtained from patients who were diagnosed and received surgical resection at the Eastern Hepatobiliary Surgery Hospital,Shanghai,China.No patients received any preoperative anticancer treatment.Patients were recruited from 2006 to 2015,and detailed information was shown in Supplementary TableS1 and TableS2.All research complied with the principles of the Declaration of Helsinki.Patient samples were obtained following informed consent according to an established protocol approved by the Ethics Committee of Eastern Hepatobiliary Surgery Hospital.Microarray analysis Arraystar Human circRNA array was adopted for profiling the circRNAs expression according to the manufacturers instructions.Agilent human mRNA array was adopted for detection of mRNA expression.Exiqon human microRNA array was adopted for detection of Accepted ArticleThis article is protected by copyright.All rights reserved.microRNA expression.All the microarray analysis was performed by KangChen Bio-tech(Shanghai,China).The expression microarrays data are available in GEO database(www.ncbi.nlm.nih.gov/geo/)with accession number GSE121815.Permission was obtained from the Ethics Committee of Eastern Hepatobiliary Surgery Hospital to submit the data to GEO.Statistical Analysis All statistical analyses were performed with SPSS 18.0 software.Qualitative variables were analyzed by chi-square test or fishers exact test.For continuous variables,if which obey the normal distribution,student t test is used to compare the differences.Otherwise,variables were compared using nonparametric test for which with an abnormal distribution.Differences between groups were compared using analysis of variance(ANOVA)when applicable or a nonparametric test.Correlation analysis was performed using the Pearson correlation coefficient method.ROC curve analysis was performed for estimate the diagnostic sensitivity and specificity.Unless otherwise specified,the results are presented as the means standard deviation(SD).All statistical tests were 2 sided,and P 0.05 was considered statistically significant.Accepted ArticleThis article is protected by copyright.All rights reserved.RESULTS Genome-wide analysis of mRNA,circRNA and microRNA expression profiles forming a noncoding regulatory RNAs network in early HCC To systematically identify the regulatory networks of coding RNAs and noncoding RNAs in the early stages HCC,ten pairs of BCLC stage 0 HCC tissues(Table S1)and nontumorous adjacent tissues were multidimensionally analyzed via microarray.The differential expression of mRNAs,circRNAs and microRNAs across the genome is shown in Figure 1A.Overall,92 circRNAs were up-regulated and 85 circRNAs were down-regulated,310 microRNAs were up-regulated and 142 microRNAs were down-regulated,9470 mRNAs were up-regulated and 10765 mRNAs were down-regulated,in the HCC tissues respectively.Except the Y chromosome,each chromosome produced abnormally expressed circRNAs(Figure 1A).In addition to the microarray results and bioinformatics annotation,a circRNA-mRNA interaction analysis was performed to identify the noncoding regulatory RNAs network that acts in the very early stage HCC(Figure 1B).According to the characteristics of ceRNA regulation(16),we adopted three criteria to identify candidate circRNA-mRNA interaction pairs:(1)We included pairs where the candidate circRNA and the mRNA shared common MREs.(2)We included pairs where the expression trends of the candidate circRNA and the mRNA were the same in both tumor tissues and nontumorous adjacent tissues.(3)To eliminate the“on-off”effect of microRNAs,we excluded circRNA-mRNA interaction pairs that were modulated by possible active microRNAs,which were differentially expressed Accepted ArticleThis article is protected by copyright.All rights reserved.between the tumor and adjacent tissues(Figure 1B).After this three-step process,126“Up-Up”RNA interaction pairs(circRNA and interacting mRNA were both upregulated in tumor tissues)and 74“Down-Down”RNA interaction pairs were identified(Figure 1C-1D and Figure S1A-S1C).These RNA pairs interact with each other through common modulatory microRNAs and represent a complex regulatory network in the very early stage HCC.The connecting line in the core of circles in Figure 1A showed the complex interaction between candidate circRNAs and mRNAs.Then,we were intrigued by the noncoding regulatory RNAs network centered on Circ-CDYL,which was the most significantly upregulated circRNA in tumor tissues.Circ-CDYL was found to interact with hepatoma-derived growth factor(HDGF)and hypoxia inducible factor asparagine hydroxylase(HIF1AN)by acts as the sponge of miR-892a and miR-328-3p,respectively(Figure 1D).We confirmed t