DataSheet1_Identification of lncRNA–miRNA–mRNA Networks Linked to Non-small Lung Cancer Resistance to Inhibitors of Epidermal Growth Factor Receptor.docx

Background: Tyrosine kinase inhibitors that act against epidermal growth factor receptor (EGFR) show strong efficacy against non-small cell lung cancer (NSCLC) involving mutated EGFRs. However, most such patients eventually develop resistance to EGFR-TKIs. Numerous researches have reported that messenger RNAs (mRNAs) and non-coding RNAs (ncRNAs) may be involved in EGFR-TKI resistance, but the comprehensive expression profile and competitive endogenous RNA (ceRNA) regulatory network between mRNAs and ncRNAs in EGFR-TKI resistance of NSCLC are incompletely known. We aimed to define a ceRNA regulatory network linking mRNAs and non-coding RNAs that may mediate this resistance.
Methods: Using datasets GSE83666, GSE75309 and GSE103352 from the Gene Expression Omnibus, we identified long non-coding RNAs (lncRNAs), microRNAs (miRNAs) and mRNAs differentially expressed between NSCLC cells that were sensitive or resistant to EGFR-TKIs. The potential biological functions of the corresponding differentially expressed genes were analyzed based KEGG pathways. We combined interactions among lncRNAs, miRNAs and mRNAs in the RNAInter database with KEGG pathways to generate transcriptional regulatory ceRNA networks associated with NSCLC resistance to EGFR-TKIs. Kaplan-Meier analysis was used to assess the ability of core ceRNA regulatory sub-networks to predict the progression-free interval and overall survival of NSCLC. The expression of two core ceRNA regulatory sub-networks in NSCLC was validated by quantitative real-time PCR.
Results: We identified 8,989 lncRNAs, 1,083 miRNAs and 3,191 mRNAs that were differentially expressed between patients who were sensitive or resistant to the inhibitors. These DEGs were linked to 968 biological processes and 31 KEGG pathways. Pearson analysis of correlations among the DEGs of lncRNAs, miRNAs and mRNAs identified 12 core ceRNA regulatory sub-networks associated with resistance to EGFR-TKIs. The two lncRNAs ABTB1 and NPTN with the hsa-miR-150–5p and mRNA SERPINE1 were significantly associated with resistance to EGFR-TKIs and survival in NSCLC. These lncRNAs and the miRNA were found to be down-regulated, and the mRNA up-regulated, in a resistant NSCLC cell line relative to the corresponding sensitive cells.
Conclusion: In this study, we provide new insights into the pathogenesis of NSCLC and the emergence of resistance to EGFR-TKIs, based on a lncRNA-miRNA-mRNA network.

研究背景：靶向表皮生长因子受体（epidermal growth factor receptor, EGFR）的酪氨酸激酶抑制剂，对携带突变型EGFR的非小细胞肺癌（non-small cell lung cancer, NSCLC）具有显著临床疗效。然而，此类患者大多最终会对EGFR-TKIs产生获得性耐药。已有大量研究证实，信使RNA（messenger RNAs, mRNAs）与非编码RNA（non-coding RNAs, ncRNAs）可能参与EGFR-TKI耐药的调控过程，但非小细胞肺癌EGFR-TKI耐药相关的mRNA与ncRNA之间的全面表达谱及内源竞争RNA（competitive endogenous RNA, ceRNA）调控网络仍未被完全阐明。本研究旨在构建可介导该耐药过程的mRNA与非编码RNA的ceRNA调控网络。
研究方法：本研究从基因表达综合数据库（Gene Expression Omnibus, GEO）中获取GSE83666、GSE75309及GSE103352数据集，筛选出EGFR-TKI敏感与耐药非小细胞肺癌细胞间的差异表达长链非编码RNA（long non-coding RNAs, lncRNAs）、微小RNA（microRNAs, miRNAs）及mRNA。基于京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes, KEGG）通路，对差异表达基因的潜在生物学功能进行富集分析。结合RNAInter数据库中lncRNAs、miRNAs与mRNAs的相互作用关系及KEGG通路富集结果，构建与非小细胞肺癌EGFR-TKI耐药相关的转录调控ceRNA网络。采用Kaplan-Meier分析法评估核心ceRNA调控子网对非小细胞肺癌患者无进展生存期与总生存期的预测价值。通过实时定量聚合酶链式反应（quantitative real-time PCR, qRT-PCR）验证两个核心ceRNA调控子网在非小细胞肺癌中的表达水平。
研究结果：本研究共筛选得到8989个差异表达lncRNAs、1083个差异表达miRNAs及3191个差异表达mRNAs。这些差异表达基因（differentially expressed genes, DEGs）共涉及968个生物学过程及31条KEGG通路。通过对lncRNAs、miRNAs及mRNAs的差异表达基因进行Pearson相关性分析，共识别出12个与EGFR-TKI耐药显著相关的核心ceRNA调控子网。其中，结合hsa-miR-150–5p的两条lncRNAs ABTB1与NPTN，以及靶mRNA SERPINE1，与非小细胞肺癌EGFR-TKI耐药及患者生存预后显著相关。相较于对应EGFR-TKI敏感细胞株，耐药非小细胞肺癌细胞株中上述lncRNAs及miRNA呈低表达状态，而靶mRNA SERPINE1呈高表达状态。
研究结论：本研究基于lncRNA-miRNA-mRNA调控网络，为非小细胞肺癌的发病机制及EGFR-TKI获得性耐药的产生提供了全新的研究视角与理论依据。