基于高通量测序对白癜风黑素细胞的综合生物信息学分析

Comprehensive bioinformatics analysis of vitiligo melanocytes based on high-throughput sequencing

目的筛选人类白癜风黑素细胞中差异表达的编码和非编码RNAs,分析它们之间相互作用并对鉴定出的重要RNAs进行表达量验证,以探讨其在白癜风发病过程中可能扮演的角色。方法对人类白癜风黑素细胞株和正常黑素细胞株进行高通量测序,用R软件进行差异表达(differentially expressed,DE)分析,利用互作基因的检索工具STRING构建蛋白相互作用网络(protein-protein interaction,PPI),用生物信息学算法miRanda鉴定miRNA-mRNA和miRNA-lncRNA的相互作用环。并用Cytoscape软件构建CeRNA网络,运用DAVID数据库对筛选获得的DE mRNAs进行功能富集和通路分析,最后对鉴定的关键RNAs进行RT-qPCR验证。结果差异表达分析共筛选出5958个DE mRNAs,1650个DE LncRNAs和450个DE miRNAs。PPI网络中top10枢纽基因(JUN,HIST2H3PS2,SOX2,EGR1,MMP9,NGF,BMP2,HSPA5,EDN1,SERPINE1)显示与白癜风发病密切相关。本研究还挖掘了与top20枢纽基因有相互作用的LncRNAs,通过取它们的交集鉴定出7个关键LncRNAs,其中分子功能和信号通路富集最丰富的3个LncRNAs(ENST0000623111、NONHSAT245258.1和NONHSAT 189782.1)被分别构建了竞争性内源性RNA(CeRNA)子网络,CeRNA子网络中预测的hsa-miR-92b-5p在白癜风黑素细胞中的表达有显著差异。另外,RT-qPCR结果与我们分析结果基本一致。结论本研究经过层层鉴定及筛选出的枢纽基因以及关键的ncRNAs可能在白癜风发病中发挥重要作用,这些发现可能为白癜风的研究提供了新的见解和策略。

Objective To screen the differentially expressed(DE)coding and non-coding RNAs in human vitiligo melanocytes and analyze their interactions and verify the expression of the identified key RNAs in order to explore their possible role in the pathogenesis of vitiligo.Methods High throughput sequencing was performed on human vitiligo melanocyte lines and normal human melanocyte lines.DE analysis was performed by R software.Protein-protein interaction(PPI)network was constructed by using the search tool STRING.The interaction loop of miRNA-mRNA and miRNA-lncRNA was identified by bioinformatics algorithm miRanda.The competitive endogenous RNA(CeRNA)network was constructed by Cytoscape software,and the functional enrichment and pathway analysis of DE mRNAs were performed by David database.Finally,the identified key RNAs were verified by Q-PCR.Results A total of 5958 DE mRNAs,1650 DE lncRNAs and 450 DE miRNAs were screened out by DE analysis.The top10 hub genes(JUN,HIST2H3PS2,SOX2,EGR1,MMP9,NGF,BMP2,HSPA5,EDN1 and SERPINE1)in the PPI network are indicated closely related to the pathogenesis of vitiligo.Furthermore,7 key lncRNAs that interact with top 20 hub genes are found.Among them,3 lncRNAs(ENST0000623111,NONHSAT245258.1 and NONHSAT189782.1)with the most abundant molecular functions and signaling pathways were constructed into CeRNA subnetwork,and the expression of hsa-miR-92b-5p predicted in the CeRNA sub-network is significantly different in vitiligo melanocytes.In addition,the results of Q-PCR were basically consistent with our analysis.Conclusion The identified hub genes and non-coding RNAs in our study may play important roles in the pathogenesis of vitiligo.Our findings provide novel insights and strategies for researches on vitiligo.