Identification of miRNAs, mRNAs, lncRNAs, and circRNAs associated with hepatocellular carcinoma recurrence after interferon treatment

AbstractObjective: Interferons have shown varied sensitivity in the treatment of hepatocellular carcinoma (HCC). Here, we studied the molecular mechanism of an interferon to explore its specific targets in the anti-HCC metastasis process and the biomarkers that can effectively predict its clinical efficacy. Methods: Differentially expressed (DE)-miRNAs, -mRNAs, -lncRNAs, and -circRNAs were screened from 15 HCC patients: five patients had received no postoperative interferon-alpha treatment (control) and 10 patients had been injected with interferon-alpha postoperatively. Tumor recurrence and metastasis occurred in five of the ten patients who had received postoperative treatment (case2), while no such symptoms occurred in the five other patients (case1). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed to determine the principal functions of the differentially expressed genes. Analyses of the networks of protein-protein interactions (PPI) and competing endogenous RNA (ceRNA) were also performed. In addition, a series-cluster analysis was performed to analyze changes in gene expression across different stages of HCC. Results: A total of 36 union miRNAs (11 up-regulated, 25 down-regulated), 175 union mRNAs (94 up-regulated, 81 down-regulated), 65 union lncRNAs (9 up-regulated, 56 down-regulated), and 52 union circRNAs (30 up-regulated, 22 down-regulated) were obtained between the control vs case1 and case2 vs case1 groups. DE-mRNAs were mainly enriched in the mitochondrial inner membrane, endocytic vesicle membrane, regulation of transcription, and DNA-templated synthesis. DE-circRNAs were mainly enriched in the Golgi apparatus, unfolded protein binding, and membrane, clathrin-coated vesicles, and endocytic recycling. The proteins that had the 11 highest values of degree were CD19, SRC, CD22, CD68, CXCL10, PAX5, BLK, CD1D, CXCR5, MS4A1, and TYMS. The integrated ceRNA network contained 162 nodes, which included 68 DE-mRNAs, 26 DE-miRNAs, 45 DE-lncRNAs, and 23 DE-circRNAs. A total of 4 DE-miRNAs, 175 DE-mRNAs, 65 DE-lncRNAs, and 52 DE-circRNAs were classified into eight profiles, respectively. Conclusion: We screened a number of mRNAs, miRNAs, lncRNAs, and circRNAs that might be the target of interferon-alpha therapy for HCC. The results may lay a foundation for investigating the different sensitivities of interferon-alpha in the treatment of HCC.

摘要：
目的：干扰素在肝细胞癌（hepatocellular carcinoma, HCC）的治疗中展现出各异的敏感性。本研究旨在探究某类干扰素抗肝细胞癌转移的分子机制，明确其作用靶点，并筛选可有效预测其临床疗效的生物标志物。
方法：本研究从15例肝细胞癌患者中筛选差异表达（differentially expressed, DE）微小RNA（miRNA）、信使RNA（mRNA）、长链非编码RNA（lncRNA）及环状RNA（circRNA）：其中5例患者术后未接受干扰素-α（interferon-α）治疗（对照组），10例患者术后接受干扰素-α注射治疗。在接受术后治疗的10例患者中，5例出现肿瘤复发与转移（病例2组），其余5例未出现上述症状（病例1组）。本研究通过基因本体（Gene Ontology, GO）及京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes, KEGG）通路富集分析，明确差异表达基因的主要功能；同时开展蛋白质相互作用（protein-protein interactions, PPI）网络与内源竞争RNA（competing endogenous RNA, ceRNA）网络分析；此外，通过时序聚类分析探究肝细胞癌不同分期的基因表达变化。
结果：对照组与病例1组、病例2组与病例1组的比较中共筛选得到36个共同差异miRNA（11个上调，25个下调）、175个共同差异mRNA（94个上调，81个下调）、65个共同差异lncRNA（9个上调，56个下调）及52个共同差异circRNA（30个上调，22个下调）。差异表达mRNA主要富集于线粒体内膜、内吞囊泡膜、转录调控及DNA模板合成过程；差异表达circRNA主要富集于高尔基体、未折叠蛋白结合、膜结构、网格蛋白包被囊泡及内吞循环过程。节点度排名前11的蛋白质分别为CD19、SRC、CD22、CD68、CXCL10、PAX5、BLK、CD1D、CXCR5、MS4A1及TYMS。整合后的ceRNA网络共包含162个节点，其中包括68个差异mRNA、26个差异miRNA、45个差异lncRNA及23个差异circRNA。共计4个差异miRNA、175个差异mRNA、65个差异lncRNA及52个差异circRNA可被分别划分为8个表达谱。
结论：本研究筛选得到若干可能作为肝细胞癌干扰素-α治疗靶点的mRNA、miRNA、lncRNA及circRNA，本研究结果可为探究干扰素-α在肝细胞癌治疗中的疗效差异提供理论基础。