Search Advanced

Message Board

Dear readers, authors and reviewers,you can add a message on this page. We will reply to you as soon as possible!

2021 Volume 46 Issue 11
Article Contents
Previous Article Next Article

QIAO Feng, ZHANG Xu. On Construction and Comprehensive Analysis of A Competitive Endogenous RNAs Network for Endometrial Carcinoma[J]. Journal of Southwest China Normal University(Natural Science Edition), 2021, 46(11): 15-22. doi: 10.13718/j.cnki.xsxb.2021.11.003
Citation: QIAO Feng, ZHANG Xu. On Construction and Comprehensive Analysis of A Competitive Endogenous RNAs Network for Endometrial Carcinoma[J]. Journal of Southwest China Normal University(Natural Science Edition), 2021, 46(11): 15-22. doi: 10.13718/j.cnki.xsxb.2021.11.003
shu

On Construction and Comprehensive Analysis of A Competitive Endogenous RNAs Network for Endometrial Carcinoma

  • 1.

    School of Mathematics and Statistics, Southwest University, Chongqing 400715, China

  • 2.

    The Science Cite Secondary School Affiliated to Chongqing Normal University, Chongqing 401331, China

More Information
  • Corresponding author: ZHANG Xu
  • Received Date: 08/09/2020
    Available Online: 20/11/2021

  • MSC: Q522

  • This study is based on the dataset of endometrial cancer gene expression from The Cancer Genome Atlas (TCGA) database. The abnormal expression of 1906 mRNAs, 753 lncRNAs and 56 miRNAs in EC samples were identified by gene differential expression analysis. Then 1906 differentially expressed mRNAs were analyzed by GO function enrichment analysis and KEGG pathway analysis. Based on the identified differentially expressed genes, a competitive endogenous RNA (ceRNA) regulatory network containing 66 mRNAs, 16 miRNAs and 37 lncRNAs was constructed. In this network, the expression levels of 16 mRNAs, 5 lncRNAs and 1 miRNAwere closely related to the overall survival rate of EC patients (P.value < 0.01). By multivariate Cox regression analysis, we constructed a risk score system containing 5lncRNAs, which has good identification and prediction ability for the survival of EC patients. In this study, the molecular interaction mechanism of EC was deeply exploredand the scope of targeting lncRNAsfurther reduced, which will be helpful for the early diagnosis, prognosis and new treatment strategy of EC.
  • 加载中
  • [1] SIEGEL R L, MILLER K D, JEMAL A. Cancer Statistics, 2019[J]. CaA Cancer Journal for Clinicians, 2019, 69(1): 7-34. doi: 10.3322/caac.21551

    CrossRef Google Scholar

    [2] SIEGELR L, MILLER K D, JEMAL A. Cancer Statistics, 2020 [J]. CA: A Cancer Journal for Clinicians, 2020, 70(1): 7-30. doi: 10.3322/caac.21590

    CrossRef Google Scholar

    [3] CHEN W, ZHENG R, BAADEP D, et al. Cancer Statistics in China, 2015 [J]. CA: A Cancer Journal for Clinicians, 2016, 66(2): 115-132. doi: 10.3322/caac.21338

    CrossRef Google Scholar

    [4] CHAUDHRY P, ASSELIN E. Resistance toChemotherapy and Hormone Therapy in Endometrial Cancer [J]. Endocrine-Related Cancer, 2009, 16(2): 363-380. doi: 10.1677/ERC-08-0266

    CrossRef Google Scholar

    [5] PONTINGC P, OLIVER P L, REIK W. Evolution and Functions of Long Noncoding RNAs [J]. Cell, 2009, 136(4): 629-641. doi: 10.1016/j.cell.2009.02.006

    CrossRef Google Scholar

    [6] JU C, LIU R F, ZHANG Y W, et al. MesenchymalStem Cell-Associated lncRNA in OsteogenicDifferentiation [J]. Biomedicine & Pharmacotherapy, 2019, 115: 108912.

    Google Scholar

    [7] KOIRALA P, HUANG J, HOT T, et al. LncRNA AK023948 is a Positive Regulator of AKT [J]. Nature Communications, 2017, 8: 14422. doi: 10.1038/ncomms14422

    CrossRef Google Scholar

    [8] BA Z W, GU LL, HAO S N, et al. Downregulation of lncRNA CASC2 Facilitates Osteosarcoma Growth and Invasion through MiR-181a [J]. Cell Proliferation, 2018, 51(1): e12409.

    Google Scholar

    [9] BESTERA C, LEE J D, CHAVEZ A, et al. An Integrated Genome-Wide CRISPRa Approach to FunctionalizelncRNAs in Drug Resistance [J]. Cell, 2018, 173(3): 649-664. e20. doi: 10.1016/j.cell.2018.03.052

    CrossRef Google Scholar

    [10] CASTELLANOS-RUBIOA, FERNANDEZ-JIMENEZN, KRATCHMAROVR, et al. A Long Noncoding RNA Associated with Susceptibility to Celiac Disease [J]. Science, 2016, 352(6281): 91-95. doi: 10.1126/science.aad0467

    CrossRef Google Scholar

    [11] ZHANG L, YAN J, LIU Q, et al. LncRNA Rik-203Contributes to Anesthesia Neurotoxicity via MicroRNA-101a-3p and GSK-3β-Mediated Neural Differentiation [J]. Scientific Reports, 2019, 9(1): 6822. doi: 10.1038/s41598-019-42991-4

    CrossRef Google Scholar

    [12] GE X S, CHEN Y B, LIAO X Y, et al. Overexpression of Long Noncoding RNA PCAT-1 is a Novel Biomarker of Poor Prognosis in Patients with Colorectal Cancer [J]. Medical Oncology, 2013, 30(2): 1-6.

    Google Scholar

    [13] YANG L, ZHANG J, JIANG A L, et al. Expression Profile of Long Non-Coding RNAs is Altered in Endometrial Cancer [J]. International Journal of Clinical and Experimental Medicine, 2015, 8(4): 5010-5021.

    Google Scholar

    [14] RAZAVI M, GHORBIAN S. Up-Regulation of Long Non-Coding RNA-PCAT-1 Promotes Invasion and Metastasis in Esophageal Squamous Cell Carcinoma [J]. EXCLI Journal, 2019, 18: 422-428.

    Google Scholar

    [15] GIBB E A, VUCIC E A, ENFIELD K S S, et al. Human Cancer Long Non-Coding RNA Transcriptomes [J]. PLoS One, 2011, 6(10): e25915. doi: 10.1371/journal.pone.0025915

    CrossRef Google Scholar

    [16] QUJ, LIUB, LIB, et al. TRIB3 Suppresses Proliferation and Invasion and Promotes Apoptosis of Endometrial Cancer CellsbyRegulatingthe AKT Signaling Pathway [J]. OncoTargetsand Therapy, 2019, 12: 2235-2245. doi: 10.2147/OTT.S189001

    CrossRef Google Scholar

    [17] SALMENA L, POLISENO L, TAY Y, et al. A ceRNA Hypothesis: The Rosetta Stone of a Hidden RNA Language? [J]. Cell, 2011, 146(3): 353-358. doi: 10.1016/j.cell.2011.07.014

    CrossRef Google Scholar

    [18] 侯丽云, 张旭, 吴珍. 多种归一化方法对miRNA微阵列数据的作用分析及比较[J]. 西南师范大学学报(自然科学版), 2020, 45(5): 98-102.

    Google Scholar

    [19] HUANG DA W, SHERMAN B T, LEMPICKI R A. Systematic and Integrative Analysis of Large Gene Lists Using DAVID Bioinformatics Resources [J]. Nature Protocols, 2009, 4(1): 44-57. doi: 10.1038/nprot.2008.211

    CrossRef Google Scholar

    [20] HUANG D W, SHERMAN B T, LEMPICKI R A. Bioinformatics Enrichment Tools: Paths Toward the Comprehensive Functional Analysis of Large Gene Lists [J]. Nucleic Acids Research, 2009, 37(1): 1-13. doi: 10.1093/nar/gkn923

    CrossRef Google Scholar

    [21] JEGGARI A, MARKS D S, LARSSON E. MiRcode: a Map of Putative microRNA Target Sites in the Long Non-Coding Transcriptome [J]. Bioinformatics, 2012, 28(15): 2062-2063. doi: 10.1093/bioinformatics/bts344

    CrossRef Google Scholar

    [22] CHOU C H, SHRESTHA S, YANG C D, et al. MiRTarBaseUpdate 2018: a Resource for Experimentally Validated MicroRNA-Target Interactions [J]. Nucleic Acids Research, 2018, 46(D1): D296-D302. doi: 10.1093/nar/gkx1067

    CrossRef Google Scholar

    [23] LIU W J, WANG X W. Prediction of Functional microRNA Targets by Integrative Modeling of microRNA Binding and Target Expression Data [J]. Genome Biology, 2019, 20(1): 1-10. doi: 10.1186/s13059-018-1612-0

    CrossRef Google Scholar

    [24] CHEN Y H, WANG X W. MiRDB: an Online Database for Prediction of Functional microRNA Targets [J]. Nucleic Acids Research, 2020, 48(D1): D127-D131. doi: 10.1093/nar/gkz757

    CrossRef Google Scholar

    [25] AGARWAL V, BELL GW, NAM JW, et al. Predicting Effective microRNA Target Sites in Mammalian mRNAs[J]. eLifeSciences, 2015, 4: e05005.

    Google Scholar

    [26] SHANNON P, MARKIEL A, OZIER O, et al. Cytoscape: aSoftware Environment for Integrated Models of BiomolecularInteraction Networks [J]. Genome Research, 2003, 13(11): 2498-2504. doi: 10.1101/gr.1239303

    CrossRef Google Scholar

    [27] BRAUNM M, OVERBEEK-WAGER E A, GRUMBO R J. Diagnosis and Management of Endometrial Cancer [J]. American Family Physician, 2016, 93(6): 468-474.

    Google Scholar

    [28] SOROSKY J I. Endometrial Cancer [J]. Obstetrics & Gynecology, 2012, 120(2, Part 1): 383-397.

    Google Scholar

  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(5)  /  Tables(1)

Export Citation
PDF
XML

Article Metrics

Article views(1101) PDF downloads(97) Cited by(0)

Access History

Other Articles By Authors

On Construction and Comprehensive Analysis of A Competitive Endogenous RNAs Network for Endometrial Carcinoma

    Corresponding author: ZHANG Xu
  • Received Date: 08/09/2020
  • Available Online: 20/11/2021

Abstract: This study is based on the dataset of endometrial cancer gene expression from The Cancer Genome Atlas (TCGA) database. The abnormal expression of 1906 mRNAs, 753 lncRNAs and 56 miRNAs in EC samples were identified by gene differential expression analysis. Then 1906 differentially expressed mRNAs were analyzed by GO function enrichment analysis and KEGG pathway analysis. Based on the identified differentially expressed genes, a competitive endogenous RNA (ceRNA) regulatory network containing 66 mRNAs, 16 miRNAs and 37 lncRNAs was constructed. In this network, the expression levels of 16 mRNAs, 5 lncRNAs and 1 miRNAwere closely related to the overall survival rate of EC patients (P.value < 0.01). By multivariate Cox regression analysis, we constructed a risk score system containing 5lncRNAs, which has good identification and prediction ability for the survival of EC patients. In this study, the molecular interaction mechanism of EC was deeply exploredand the scope of targeting lncRNAsfurther reduced, which will be helpful for the early diagnosis, prognosis and new treatment strategy of EC.

    HTML

  • 子宫内膜癌(EC)是世界上最常见的女性恶性肿瘤之一[1-4]. 因此，研究EC的分子发病机制，寻找与EC发生、发展和预后相关的生物标志物尤为重要.

    长非编码RNA(lncRNA)被定义为长度大于200 bp且无蛋白质编码潜力的转录物[5]. 近年来，越来越多的研究表明，lncRNA在细胞发育、分化、增殖、迁移和转移的生理和病理过程中起着重要作用[6-11]. 研究表明，某些lncRNA可作为潜在的癌症诊断生物标志物[12-14]；MALAT1可作为筛查肺癌、子宫内膜间质肉瘤、宫颈癌和肝癌的生物标志物[15]. TRIB3已被证明是EC的潜在治疗靶点，因为它可以通过调节AKT信号通路促进EC细胞凋亡和抑制EC细胞增殖和迁移[16].

    内源竞争RNA(ceRNA)假说揭示了RNA相互作用的新机制，认为信使RNA(mRNA)和其他非编码RNA可以通过共同的microRNA(miRNA)反应元件竞争性地与miRNA结合，从而调节某些基因的表达水平[17]. 近年来，越来越多的研究证实ceRNA调控理论与肿瘤的发生、发展和预后密切相关[18].

    目前，EC中的ceRNA调节网络机制尚不清楚.

    本文先对mRNA，lncRNA和miRNA分别进行差异表达分析；随后通过GO功能富集分析和KEGG通路分析，进一步挖掘了差异表达的mRNA潜在的生物学功能；接着通过成对预测，整合差异表达的mRNA(DEmRNA)、差异表达的lncRNA(DElncRNA)和差异表达的miRNA(DEmiRNA)，构建了与EC相关的ceRNA网络，帮助我们挖掘EC发生的分子机制. 为了确定与EC相关的预后因素，对ceRNA网络中的RNA进行了生存分析. 通过多变量Cox回归，构建了一个风险评分系统，对EC患者生存期具有良好的鉴别和预测能力. 本研究对理解EC的分子相互作用机制提供了新的见解，进一步缩小了靶向lncRNA的范围，也将有助于EC的早期诊断、预后及新治疗策略的制定.

1.   数据和方法

    1.1.   数据来源及处理

  • 本文所使用的数据集来自从肿瘤基因图谱(TCGA)数据库、EC的mRNA(包括lncRNA)和miRNA表达数据及相应临床数据(https://genome-cancer.ucsc.edu/，2019年7月31日更新). 下载的mRNA和miRNA表达数据分别包含583个样本(35个正常样本，548个肿瘤样本)和575个样本(33个正常样本，542个肿瘤样本). 我们用GENCODE数据库(https://www.gencodegenes.org/，版本32)以识别mRNA和lncRNA. RNA表达数据包含19 668个mRNA、14 090个lncRNA和1881个miRNA.

    • 1.2.   差异分析

  • 应用R软件中的edge软件包筛选548例肿瘤组织与35例正常组织的差异表达的mRNA和lncRNA. 用R软件中的limma软件包对542例肿瘤组织和33例正常组织样本进行了miRNA的差异表达分析[18]. 分别在两组水平分析显著异常表达的lncRNA，miRNA和mRNA：中分化至高分化(G1-G2期)EC样本与正常样本、低分化(G3-G4期)EC样本与正常样本. DEmRNA，DElncRNA和DEmiRNA的筛选标准为：假发现率(FDR) < 0.01且|log2(FC)(fold change)|＞2. 然后，用火山图显示符合标准的差异表达的lncRNA，miRNA和mRNA. 另外，通过韦恩(Venn)图显示了G1-G2期与G3-G4期两组样本中相交的异常表达基因，便于进行下游分析.

    • 1.3.   DEmRNA的功能富集分析

  • 为了探索DEmRNA的潜在生物学功能，利用Database for Annotation，Visualization and Integrated Discovery(DAVID)(https://david.ncifcrf.gov/)数据库对异常表达的基因进行GO功能富集分析和KEGG通路富集分析[19-20]. 在GO和KEGG通路分析中，P.value < 0.01被认为具有统计学意义.

    • 1.4.   构建ceRNA调控网络

  • 为了进一步理解mRNA，lncRNA和miRNA在EC中的相互作用机制，构建了基于DEmRNA，DEmiRNA和DElncRNA相互作用的ceRNA网络. DEmiRNA的靶向lncRNA是基于miRcode数据库[21](http://www.mircode.org/)进行预测. 然后，利用miRTarBase[22](http://mirtarbase.mbc.nctu.edu.tw/)，miRDB[23-24](http://www.mirdb.org/)和TargetScan[25](http://www.targetscan.org/)数据库预测miRNA靶向的mRNA. 为了获得更可靠的miRNA与mRNA的关系，利用3个数据库预测结果的交集，建立了一个lncRNA-miRNA-mRNA调控网络. 最后用Cytoscape[26](http://www.cytoscape.org/)软件可视化ceRNA网络.

    • 1.5.   生存分析

  • 用R软件中的survival软件包对ceRNA网络包含的mRNA，lncRNA和miRNA进行生存分析. 以RNA表达水平的中位数作为截止值，将患者分为高表达组和低表达组. 对数秩P.value < 0.05被认为具有统计学意义. 通过对ceRNA中的RNA进行Kaplan-Meier (K-M) 生存分析，获得了与总体生存时间相关的mRNA，lncRNA和miRNA. 另外，用R软件中的survminer软件包绘制K-M生存曲线，进一步验证mRNA，lncRNA和miRNA的预后价值.

    • 1.6.   构建风险评分系统

  • 构建lncRNA风险评分系统为子宫内膜癌患者的早期诊断提供便利. 将EC患者样本按照1：1的比例随机分为训练集和测试集，然后，基于训练集利用最大似然法建立与生存相关的DElncRNA的Cox风险比例回归模型，并计算模型的回归系数(β). 最后，构建了一个包含5个lncRNA的子宫内膜癌预后风险评分系统.

    其中：P表示预后指数(Prognostic index)，xi(i=1，2，3，4，5)分别代表WT1-AS，PRICKLE2-AS2，LINC00491，ALDH1L1-AS2和ADAMTS9-AS1的表达水平. 为了评估风险评分系统的识别和预测能力，构建了K-M生存曲线和时间依赖性受试者操作特征(ROC)曲线.

    • 1.7.   单变量和多变量Cox回归分析

  • 为了探索EC患者的临床特征，包括年龄(Age)、临床分期(Clinical stage)、组织学分级(Neoplasm histologic grade)、体重(Weight)和种族(Race)是否与总体生存率有显著相关，我们进行了单变量Cox回归分析. 然后，年龄、临床分期、组织学分级和风险评分水平作为候选变量被纳入多元Cox回归分析. P.value＜0.05被认为具有显著统计学意义，并计算各变量的风险比和95%置信区间.

2.   结果与分析

    2.1.   差异表达分析

  • 如图 1(a)所示，在G1-G2期的EC组织和正常组织样本中识别了2 548个DEmRNA(1 224个上调，1 324个下调)，1 146个DElncRNA(640个上调，506个下调)，72个DEmiRNA(13个上调，59个下调)；如图 1(b)所示，在G3-G4期的EC组织和正常组织样本中发现了2 695个DEmRNA(940个上调，1 755个下调)，1 347个DElncRNA(373个上调，974个下调)，80个DEmiRNA(14个上调，66个下调). 两组差异基因的交叉部分由753个lncRNA，58个miRNA和1 906个mRNA组成，这些被认为是早期EC发展的关键基因(图 1(c)).

    • 2.2.   DEmRNA的功能富集分析

  • 我们进一步研探究了1 960个DEmRNA的潜在生物学功能. 通过GO功能富集分析和KEGG通路分析，筛选出121个显著富集的GO术语(P.value＜0.01). 在这些术语中，“表皮发育”“端粒组织”“细胞信号”“肌肉收缩”和“依赖DNA复制的核小体组装”是前5位的GO术语(图 2(a))；确定了56条DEmRNA显著富集的KEGG通路、22条KEGG通路在P.value < 0.01时被确定为具有统计学意义，DEmRNA在“hsa04080：神经活性配体-受体相互作用”“hsa04270：血管平滑肌收缩”“hsa04020：钙信号通路”“hsa04022：cGMP-PKG信号通路”“hsa04110：细胞周期”“hsa04014：Ras信号通路”等信号通路显著富集(图 2(b)).

    • 2.3.   构建ceRNA调控网络

  • 利用miRcode数据集预测753个DElncRNA和53个DEmiRNA，成功地鉴定出136个miRNA-lncRNA对. 然后，利用TargetScan，miRDB和miRTarBase数据库，分析了58个DEmiRNA和1 906个DEmRNA，发现了84个miRNA和mRNA相互作用对. 最后构建了一个包含66个mRNA，16个miRNA和37个lncRNA的ceRNA调控网络(图 3).

    • 2.4.   生存分析

  • 为了确定与EC患者预后相关的mRNA，lncRNA和miRNA，对ceRNA中的每个RNA(66个mRNA，37个lncRNA和16个miRNA)进行了K-M生存分析和Log-Rank检验. 最后发现，16个mRNA(NR3C1，CIT，SOX11，CDC25A，RECK，AURKA，E2F1，ONECUT2，SALL3，SLC2A4，GFBP5，POLQ，RGS2，MNX1，KLF9和RRM2)，5个lncRNA(WT1-AS，PRICKLE2-AS2，ADAMTS9-AS1，ALDH1L1-AS2和LINC00491)和1个miRNA(hsa-mir-182)与EC患者的总体生存率显著相关(P.value < 0.05)(图 4).

    • 2.5.   构建与lncRNA相关的风险评分系统

  • 基于与总体生存率显著相关的5个lncRNA，应用多元Cox回归分析来构建风险评分系统，其贡献由其相关系数加权，最终的风险评分公式为：

    其中：P表示预后指数(Prognostic index)，xi(i=1，2，3，4，5)分别代表WT1-AS，PRICKLE2-AS2，LINC00491，ALDH1L1- AS2和ADAMTS9-AS1的表达水平. 风险评分大于最佳截断值0.945的患者被视为高危患者(212名患者)，而风险评分小于或等于0.945的患者被视为低危患者(308名患者). 特别的，根据K-M和时间依赖性ROC曲线分析，这两个组的设计均提高了对子宫内膜癌高、低危患者的预测正确率(图 5(b)，(c)). 基因表达热图和患者评分散点图(图 5(a))显示了520例EC患者生存期的5个lncRNA表达谱和风险评分以及垂直虚线0.945的风险评分的截止值. 采用单因素Cox回归分析筛选520例临床资料完整的EC患者总体生存率相关的特征，结果表明，年龄、临床分期、组织学分级的预后价值具有统计学意义. 在多因素Cox回归分析中，年龄、临床分期、组织学分级和风险评分与EC患者的预后密切相关. 因此，我们构建的lncRNA风险评分系统可以作为子宫内膜癌患者的独立预后指标(表 1).

3.   讨论

  • 子宫内膜癌是一种致命的女性恶性肿瘤. 在过去的20年里，EC死亡率翻了一番. 只有20%的EC患者在绝经前被诊断出来[27-28]. 因此，研究EC的分子发病机制，寻找与EC发生、发展和预后相关的生物标志物尤为重要. 我们首先对收集到的EC患者的样本，分别对mRNA，lncRNA和miRNA数据进行了差异表达分析，最终得到在EC组织中异常表达的mRNA，lncRNA和miRNA. 然后，在此基础上通过成对预测获得了miRNA的靶向lncRNA和mRNA. 最终构建了EC的一个ceRNA调控网络进一步去理解EC分子间相互作用机制. 接下来，通过对包含在ceRNA网络中的RNA进行了生存分析，筛选出了与EC患者总体生存显著相关的mRNA，lncRNA和miRNA. 最后，根据筛选出的5个与EC患者总体生存时间密切相关的lncRNA，利用Cox多元回归构建了一个风险评分系统. K-M生存曲线和时间依赖性ROC曲线进一步验证了该风险评分系统具备良好的预测能力，有助于子宫内膜癌早期诊断. 单因素Cox回归和多因素Cox回归分析的结果也证明了这个风险评分可以作为EC患者生存时间的独立预后指标. 本研究深入挖掘了EC的分子相互作用机制并进一步缩小了靶向lncRNA的范围，将有助于EC的早期诊断、预后及新的治疗策略的制定. 由于缺乏其它类似的EC相关lncRNA数据库，因此未进行外部验证.

Figure (5)  Table (1) Reference (28)

Catalog

    Copyright Journal of Southwest China Normal University(Natural Science Edition)

    Address： Tel：023-68252540，68254576，68254478 E-mail：xbbjb@swu.edu.cn

    Supported by:Beijing Renhe Information Technology Co. Ltd

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return