与喉乳头状瘤恶变进程及预后相关分子标志物研究

Bioinformatics analysis of key molecular markers for malignant transformation of laryngeal papilloma

目的 基于生物信息学分析筛选影响喉乳头状瘤恶变进程及预后的分子标志物。方法 从GEO和TCGA数据库分别下载成人喉乳头状瘤数据集GSE10935和喉鳞癌转录组数据,使用R语言筛选两组数据差异表达基因(DEGs)。利用韦恩图分析筛选两组共同DEGs后,使用GEPIA数据库绘制Kaplan-Meier生存曲线进行生存分析,筛选出候选基因。利用HPA数据库分析候选基因的蛋白表达情况得到关键基因后,进行单/多因素COX回归分析,及GO、KEGG功能富集分析。结果 从GSE10935数据集筛选出112个与喉乳头状瘤发生发展相关的DEGs,从喉鳞癌转录组数据筛选出1817个与喉鳞癌发生发展相关的DEGs。通过韦恩图分析得到共同DEGs 24个。GEPIA在线网站分析显示与正常组织相比,FSCN1、MMP1、IFI27在头颈鳞癌组织(HNSCC)中高表达,ALDH3A1、HLF、MMRN1低表达,差异均有统计学意义(FSCN1:P=0.002 9、MMP1:P=0.047、IFI27:P=0.035、ALDH3A1:P=0.024、HLF:P=0.008、MMRN1:P=0.036)。生存分析显示FSCN1、MMP1、IFI27低表达组患者预后明显优于高表达组患者,ALDH3A1、HLF、MMRN1高表达组患者预后明显优于低表达组患者。HPA数据库分析候选基因免疫组化结果发现,FSCN1蛋白在HNSCC中显著高表达,而ALDH3A1蛋白低表达。进一步生存分析表明FSCN1和ALDH3A1的表达水平是影响HPV相关的HNSCC患者预后的独立风险因素。结论 FSCN1和ALDH3A1可能是影响喉乳头状瘤恶变的关键基因,FSCN1高表达、ALDH3A1低表达显著影响HPV相关的HNSCC患者预后,可能是潜在阻滞喉乳头状瘤恶变进程的分子靶标。

Objective To screen for molecular markers affecting malignant transformation and prognosis of laryngeal papilloma using bioinformatics analysis. Methods The GSE10935 gene expression profile of adult laryngeal papilloma was downloaded from the Gene Expression Omnibus database, and the transcriptome data for laryngeal squamous cell carcinoma were downloaded from The Cancer Genome Atlas. Differentially expressed genes(DEGs) in each dataset were identified using limma and DESeq2 R package. Venn diagrams analysis was conducted for identifying common DEGs. Survival analysis was performed by plotting the Kaplan-Meier curves in the Gene Expression Profiling Interactive Analysis(GEPIA) database to screen for candidate genes. Protein expression in the Human Protein Atlas database was analyzed to identify key genes. Univariate/multivariate Cox regression analysis and functional enrichment analyses were performed using Gene Ontology and Kyoto Encyclopedia of Genes and Genomes. Results A total of 112 DEGs were related to the occurrence and development of laryngeal papilloma, and 1817 DEGs were related to laryngeal squamous cell carcinoma. Twenty-four common DEGs were identified using Venn diagram analysis. GEPIA revealed that the expression of FSCN1, MMP1, and IFI27 was upregulated, while that of ALDH3A1, HLF, and MMRN1 was downregulated in head and neck squamous cell carcinoma(HNSCC) samples compared with that in normal tissue samples;all differences were significant(FSCN1:P=0.002 9, MMP1:P=0.047, IFI27:P=0.035, ALDH3A1:P=0.024, HLF:P=0.008, MMRN1:P=0.036). Survival analysis revealed that the overexpression of FSCN1, MMP1, and IFI27 and low expression of ALDH3A1, HLF, and MMRN1 affected overall survival. Immunohistochemical analysis showed high FSCN1 expression and low ALDH3A1 expression in HNSCC samples. Further survival analysis showed that the expression levels of FSCN1 and ALDH3A1 were independent risk factors affecting the prognosis of human papillomavirus-related HNSCC. Conclusion FSCN1 and ALDH3A1 are plausible key genes in the malignant transformation of laryngeal papilloma. High FSCN1 expression and low ALDH3A1 expression affect the prognosis of human papillomavirus-related HNSCC and are potential molecular targets for suppressing malignant transformation.