第一部分  中文摘要

【背景】随着恶性肿瘤的诊断技术的不断升级，手术水平不断提高，很多肿瘤患者的生存时间逐年增加，但胆管癌（CCA）患者的远期生存率仍不理想。这些情况提示胆管癌的综合诊治可能已经到达瓶颈，需要阐明新的关键作用位点及治疗点，才能为胆管癌的治疗提供新的思路。随着对 miRNA 的研究越来越多，miRNA 在胆管癌中的作用逐渐彰显。我们期望使用生物信息学的方法寻找 CCA 中与预后相关的 miRNA 并构建预后评分，探索其临床预测意义和下游可能的靶基因，研究其影响 CCA 进展的可能机制。

【方法】从癌症基因组图谱（The Cancer Genome Atlas, TCGA）筛选与 CCA 相关的差异表达 microRNA（miRNA）和 mRNA。利用加权基因共表达网络分析（WGCNA）聚类表达模式相似的miRNA 或 mRNA，以鉴定与 CCA 高度相关的 miRNA 或 mRNA。使用Cox 比例风险回归分析和基于生存时间及状态的最小绝对收缩算子（Lasso）鉴定与整体生存（OS）相关的 miRNA。最终开发出由 3 种 miRNA 构成的 CCA 的预后标志物模型。利用在线数据库探究 3 种 miRNA 标志物的潜在下游 mRNA 靶点，使用 Kaplan-Meier方法评估目标 mRNA 对预后的影响。对目标 mRNA 进行功能富集分析，探究 miRNA-mRNA 轴在 CCA 进展过程中的作用机制。

【结果】由 3 个 miRNA（mir-129-1、mir−4524a 和 mir-380）构成的预后风险评分可以较好地评价 CCA 患者的预后。与高风险评分组的 CCA 患者相比，低风险评分组的患者具有更好的 OS 和 PFS，且两组的差异具有统计学意义。此外我们预测了这 3 个 miRNA的 39 个下游 mRNA 靶点。通过功能富集分析，我们发现这些 mRNA 主要在细胞通讯等通路富集。此外，生存分析和 Cox 比例风险回归分析提示，3 个 mRNA 靶点（KCNN1、KIF2C、DSC3）对 CCA 患者有明显的预后价值。

【结论】我们采用生物信息学方法分析了从 TCGA 获得的 CCA 数据，确定了 3 个与预后相关的 miRNA，并且探究 了其下游的潜在靶点。然而，miR-129-1、miR-4524a 和miR-380 通过靶向 mRNA 影响 CCA 肿瘤发生和发展的详细机制需要进一步探索和研究。

第二部分    中文摘要

【背景】胆管癌起源于胆管上皮，曾被认为是一种罕见的肿瘤。胆管癌是一个通用术语，用于描述肝门、远端和肝内胆管癌，它是胆道系统中最难治疗的疾病之一。研究表明，胆管癌的发病率正在增加，而且预后不佳。肝内胆管癌的 5 年生存率为 2%–15%，肝外胆管癌为 2%–30% 。本研究的目的是根据先前生物信息学分析的结果，确定 miR-380 对胆管癌细胞增殖、S 期阻滞和侵袭能力的影响，明确其下游可能影响的靶基因和细胞通路。

【方法】使用 TargetScan 数据库和双荧光素酶报告基因系统确定 LIS1 是否为 miR-380的靶基因。通过细胞计数试剂盒、流式细胞术和 Transwell 实验检测 miR-380 和 LIS1 对HCCC-9810/HuCCT1/QBC939 细胞增殖、S 期比例和侵袭性的影响。采用 Western blot 法确定 miR-380 对 MMP-2/p-AKT 的影响。

【结果】胆管癌中 miR-380 的表达降低，而 LIS1 的表达增加。确认 LIS1 为 miR-380 的靶基因。miR-380 模拟物转染抑制了 HCCC-9810/HuCCT1/QBC939 细胞的增殖、S 期阻滞和侵袭， LIS1 逆转了这些抑制效果。 miR-380 抑制剂促进了 HCCC-

9810/HuCCT1/QBC939 细胞的增殖、S 期比例和侵袭性。si-LIS1 挽救了 miR-380 抑制剂的促进作用。miR-380 的过表达抑制了 MMP-2/p-AKT/LIS1 的表达，而 miR-380 抑制剂促进了它们的表达。

【结论】miR-380 在胆管癌中表达异常，其上调通过对 LIS1 的靶向调控降低了 MMP-2/p-AKT 的表达，从而抑制了胆管癌细胞的增殖、侵袭和迁移。

PART Ⅰ   Abstract

Background With the continuous advancement of diagnostic technologies and improvement in surgical techniques in the management of malignant tumors, the survival time of many cancer patients has been increasing over the years. However, the long-term survival rate of patients with cholangiocarcinoma (CCA) remains unsatisfactory. These circumstances suggest that the comprehensive management of CCA may have reached a bottleneck, necessitating the elucidation of new key regulatory sites and therapeutic targets to provide novel insights into the treatment of CCA. As the role of miRNA in CCA becomes increasingly evident, we aimed to identify prognostically relevant miRNAs in CCA using bioinformatics approaches, construct a prognostic score, explore their clinical predictive significance, identify potential downstream target genes, and investigate the possible mechanisms underlying their impact on CCA progression. Methods Differential expression microRNAs (miRNAs) and mRNAs associated with CCA were screened from The Cancer Genome Atlas (TCGA) dataset. Weighted gene co-expression network analysis (WGCNA) was utilized to cluster miRNAs or mRNAs with similar expression patterns to identify those highly correlated with CCA. Cox proportional hazards regression analysis and least absolute shrinkage and selection operator (Lasso) based on survival time and status were employed to identify miRNAs associated with overall survival (OS). Eventually, a prognostic model composed of three miRNAs was developed for CCA. The potential downstream mRNA targets of the three miRNA biomarkers were explored using online databases, and the impact of target mRNAs on prognosis was evaluated using Kaplan-Meier methods. Functional enrichment analysis of target mRNAs was performed to investigate the mechanisms underlying the role of the miRNA-mRNA axis in CCA progression.

Results A prognostic risk score composed of three miRNAs (mir-129-1, mir-4524a, and mir-380) could effectively assess the prognosis of CCA patients. Patients in the low-risk score group had significantly better OS and progression-free survival (PFS) compared to those in the high-risk score group, with statistically significant differences observed between the two groups. Additionally, we predicted 39 downstream mRNA targets of these three miRNAs. Functional enrichment analysis revealed that these mRNAs were mainly enriched in pathways related to cell communication. Furthermore, survival analysis and Cox proportional hazards regression analysis indicated that three mRNA targets (KCNN1, KIF2C, DSC3) had significant prognostic value for CCA patients.

Conclusion Through bioinformatics analysis of CCA data obtained from TCGA, we identified three miRNAs associated with prognosis and explored their potential downstream targets. However, the detailed mechanisms by which miR-129-1, miR-4524a, and miR-380 influence the occurrence and progression of CCA through mRNA targeting require further exploration and research.

PART Ⅱ    Abstract

Background Cholangiocarcinoma originates from bile duct epithelium and was once  onsidered a rare tumor. Cholangiocarcinoma is a generic term used to describe cancers of the hepatic hilum, distal, and intrahepatic bile ducts, and it is one of the most difficult diseases to treat within the biliary system. A recent study has indicated that the incidence of cholangiocarcinoma is increasing and the prognosis is poor. The 5-year survival rate for intrahepatic cholangiocarcinoma ranges from 2% to 15%, while for extrahepatic cholangiocarcinoma it is between 2% to 30%. The aim of this study was to determine the impact of miR-380 on the proliferation, S-phase arrest, and invasive capability of cholangiocarcinoma cells, and to identify its downstream target genes and cellular pathways based on previous

bioinformatics analysis.

Methods The TargetScan database and a dual-luciferase reporter assay system were employed to ascertain whether LIS1 is a target gene of miR-380. Cell Counting Kit-8, flow cytometry, and Transwell assays were utilized to detect the effects of miR-380 and LIS1 on the proliferation, S-phase ratio, and invasiveness of HCCC-9810/HuCCT1/QBC939 cells. Western blot analysis was used to determine the influence of miR-380 on MMP-2/p-AKT.

Results The expression of miR-380 was found to be decreased in cholangiocarcinoma, while LIS1 expression was increased. LIS1 was confirmed to be a target gene of miR-380. Transfection with miR-380 mimics inhibited the proliferation, S-phase arrest, and invasion of HCCC-9810/HuCCT1/QBC939 cells, and the inhibitory effects were reversed by LIS1. The miR-380 inhibitor promoted the proliferation, S-phase ratio, and invasiveness of these cells. si-LIS1 rescued the promotive effect of the miR-380 inhibitor. Overexpression of miR-380 inhibited the expression of MMP-2/p-AKT/LIS1, whereas the miR-380 inhibitor enhanced their expression.

Conclusion Abnormal expression of miR-380 is closely related to cholangiocarcinoma, and its pregulation inhibits the proliferation, invasion, and migration of cholangiocarcinoma cells by targeting LIS1, thereby reducing the expression of MMP-2/p-AKT.