Abstract：
BACKGROUND: Uncontrolled cellular proliferation may result in the progression of diseases such as cancer that promote organism death. Programmed cell death (PCD) is an important mechanism that ensures the quality and quantity of cells, which could be developed as a potential biomarker for disease diagnosis and treatment.
METHODS: RNA-seq data and clinical information of nasopharyngeal carcinoma (NPC) patients were downloaded from the Gene Expression Omnibus (GEO), and 1548 PCD-related genes were collected. We used the \"limma\" package to analyze differentially expressed genes (DEGs). The STRING database was used for protein interaction analysis, and the least absolute shrinkage and selection operator (Lasso) and support vector machines (SVMs) regression analyses were used to identify biomarkers. Then, the timeROC package was used for classifier efficiency assessment, and the \"CIBERSORT\" package was used for immune infiltration analysis. Wound healing and transwell migration assay were performed to evaluate migration and invasion.
RESULTS: We identified 800 DEGs between our control and NPC patient groups, in which 59 genes appeared to be PCD-related DEGs, with their function closely associated with NPC progression, including activation of the PI3K-Akt, TGF-β, and IL-17 signaling pathways. Furthermore, based on the STRING database, Cytoscape and six algorithms were employed to screen 16 important genes (GAPDH, FN1, IFNG, PTGS2, CXCL1, MYC, MUC1, LTF, S100A8, CAV1, CDK4, EZH2, AURKA, IL33, S100A9, and MIF). Subsequently, two reliably characterized biomarkers, FN1 and MUC1, were obtained from the Lasso and SVM analyses. The Receiver operating characteristic (ROC) curves showed that both biomarkers had area under the curve (AUC) values higher than 0.9. Meanwhile, the enrichment analysis showed that in NPC patients, the FN1 and MUC1 expression levels correlated with programmed cell death-related pathways. The enrichment analysis and cellular experimental results indicated that FN1 and MUC1 were overexpressed in NPC cells and associated with programmed cell death-related pathways. Importantly, FN1 and MUC1 severely affected the ability of NPC cells to migrate, invade, and undergo apoptosis. Finally, medroxyprogesterone acetate and 8-Bromo-cAMP acted as drug molecules for the docking of FN1 and MUC1 molecules, respectively, and had binding capacities of -9.17 and -7.27 kcal/mol, respectively.
CONCLUSIONS: We examined the PCD-related phenotypes and screened FN1 and MUC1 as reliable biomarkers of NPC; our findings may promote the development of NPC treatment strategy.
METHODS: RNA-seq data and clinical information of nasopharyngeal carcinoma (NPC) patients were downloaded from the Gene Expression Omnibus (GEO), and 1548 PCD-related genes were collected. We used the \"limma\" package to analyze differentially expressed genes (DEGs). The STRING database was used for protein interaction analysis, and the least absolute shrinkage and selection operator (Lasso) and support vector machines (SVMs) regression analyses were used to identify biomarkers. Then, the timeROC package was used for classifier efficiency assessment, and the \"CIBERSORT\" package was used for immune infiltration analysis. Wound healing and transwell migration assay were performed to evaluate migration and invasion.
RESULTS: We identified 800 DEGs between our control and NPC patient groups, in which 59 genes appeared to be PCD-related DEGs, with their function closely associated with NPC progression, including activation of the PI3K-Akt, TGF-β, and IL-17 signaling pathways. Furthermore, based on the STRING database, Cytoscape and six algorithms were employed to screen 16 important genes (GAPDH, FN1, IFNG, PTGS2, CXCL1, MYC, MUC1, LTF, S100A8, CAV1, CDK4, EZH2, AURKA, IL33, S100A9, and MIF). Subsequently, two reliably characterized biomarkers, FN1 and MUC1, were obtained from the Lasso and SVM analyses. The Receiver operating characteristic (ROC) curves showed that both biomarkers had area under the curve (AUC) values higher than 0.9. Meanwhile, the enrichment analysis showed that in NPC patients, the FN1 and MUC1 expression levels correlated with programmed cell death-related pathways. The enrichment analysis and cellular experimental results indicated that FN1 and MUC1 were overexpressed in NPC cells and associated with programmed cell death-related pathways. Importantly, FN1 and MUC1 severely affected the ability of NPC cells to migrate, invade, and undergo apoptosis. Finally, medroxyprogesterone acetate and 8-Bromo-cAMP acted as drug molecules for the docking of FN1 and MUC1 molecules, respectively, and had binding capacities of -9.17 and -7.27 kcal/mol, respectively.
CONCLUSIONS: We examined the PCD-related phenotypes and screened FN1 and MUC1 as reliable biomarkers of NPC; our findings may promote the development of NPC treatment strategy.
摘要：
背景：不受控制的细胞增殖可能导致疾病如癌症的进展，从而促进生物体的死亡。细胞程序性死亡(PCD)是保证细胞质量和数量的重要机制,它可以被开发为疾病诊断和治疗的潜在生物标志物。
方法：从基因表达综合（GEO）下载鼻咽癌（NPC）患者的RNA-seq数据和临床信息，收集了1548个PCD相关基因。我们使用“limma”软件包分析差异表达基因(DEGs)。STRING数据库用于蛋白质相互作用分析，和最小绝对收缩和选择算子(Lasso)和支持向量机(SVMs)回归分析用于识别生物标志物。然后，timeROC软件包用于分类器效率评估，并使用“CIBERSORT”软件包进行免疫浸润分析。进行伤口愈合和transwell迁移测定以评估迁移和侵袭。
结果：我们在对照组和NPC患者之间确定了800个DEG，其中59个基因似乎是PCD相关的DEGs，它们的功能与NPC进展密切相关，包括激活PI3K-Akt，TGF-β,和IL-17信号通路。此外，基于STRING数据库，Cytoscape和六种算法用于筛选16个重要基因(GAPDH，FN1,IFNG,PTGS2,CXCL1,MYC,MUC1,LTF,S100A8,CAV1,CDK4,EZH2,AURKA,IL33、S100A9和MIF)。随后，两个可靠表征的生物标志物，FN1和MUC1是从Lasso和SVM分析中获得的。受试者工作特征(ROC)曲线显示两种生物标志物具有高于0.9的曲线下面积(AUC)值。同时,富集分析表明，在鼻咽癌患者中，FN1和MUC1表达水平与程序性细胞死亡相关通路相关。富集分析和细胞实验结果表明，FN1和MUC1在NPC细胞中过表达，并与程序性细胞死亡相关通路有关。重要的是,FN1和MUC1严重影响了NPC细胞的迁移能力，入侵,并经历凋亡。最后,醋酸甲羟孕酮和8-溴-cAMP作为药物分子，用于FN1和MUC1分子的对接，分别,并且具有-9.17和-7.27kcal/mol的结合能力，分别。
结论：我们检查了PCD相关表型，并筛选了FN1和MUC1作为NPC的可靠生物标志物；我们的发现可能促进NPC治疗策略的发展。
方法：从基因表达综合（GEO）下载鼻咽癌（NPC）患者的RNA-seq数据和临床信息，收集了1548个PCD相关基因。我们使用“limma”软件包分析差异表达基因(DEGs)。STRING数据库用于蛋白质相互作用分析，和最小绝对收缩和选择算子(Lasso)和支持向量机(SVMs)回归分析用于识别生物标志物。然后，timeROC软件包用于分类器效率评估，并使用“CIBERSORT”软件包进行免疫浸润分析。进行伤口愈合和transwell迁移测定以评估迁移和侵袭。
结果：我们在对照组和NPC患者之间确定了800个DEG，其中59个基因似乎是PCD相关的DEGs，它们的功能与NPC进展密切相关，包括激活PI3K-Akt，TGF-β,和IL-17信号通路。此外，基于STRING数据库，Cytoscape和六种算法用于筛选16个重要基因(GAPDH，FN1,IFNG,PTGS2,CXCL1,MYC,MUC1,LTF,S100A8,CAV1,CDK4,EZH2,AURKA,IL33、S100A9和MIF)。随后，两个可靠表征的生物标志物，FN1和MUC1是从Lasso和SVM分析中获得的。受试者工作特征(ROC)曲线显示两种生物标志物具有高于0.9的曲线下面积(AUC)值。同时,富集分析表明，在鼻咽癌患者中，FN1和MUC1表达水平与程序性细胞死亡相关通路相关。富集分析和细胞实验结果表明，FN1和MUC1在NPC细胞中过表达，并与程序性细胞死亡相关通路有关。重要的是,FN1和MUC1严重影响了NPC细胞的迁移能力，入侵,并经历凋亡。最后,醋酸甲羟孕酮和8-溴-cAMP作为药物分子，用于FN1和MUC1分子的对接，分别,并且具有-9.17和-7.27kcal/mol的结合能力，分别。
结论：我们检查了PCD相关表型，并筛选了FN1和MUC1作为NPC的可靠生物标志物；我们的发现可能促进NPC治疗策略的发展。
