Department of Plastic Surgery, The First Hospital of China Medical University, Shenyang, 11000 Liaoning Province, China.
Comput Math Methods Med. 2021 Dec 28;2021:9652768. doi: 10.1155/2021/9652768. eCollection 2021.
Melanoma is the deadliest type of skin cancer. Until now, its pathological mechanisms, particularly the mechanism of metastasis, remain largely unknown. Our study on the identification of genes in association with metastasis for melanoma provides a novel understanding of melanoma.
From the Gene Expression Omnibus (GEO) database, the gene expression microarray datasets GSE46517, GSE7553, and GSE8401 were downloaded. We made use of R aiming at analyzing the differentially expressed genes (DEGs) between metastatic and nonmetastatic melanoma. R was also used in differentially expressed miRNA (DEM) data mining from GSE18509, GSE19387, GSE24996, GSE34460, GSE35579, GSE36236, and GSE54492 datasets referring to Li's study. Based on the DEG and DEM data, we performed functional enrichment analysis through the application of the DAVID database. Furthermore, we constructed the protein-protein interaction (PPI) network and established functional modules by making use of the STRING database. Through making use of Cytoscape, the PPI results were visualized. We predicted the targets of the DEMs through applying TargetScan, miRanda, and PITA databases and identified the overlapping genes between DEGs and predicted targets, followed by the construction of DEM-DEG pair network. The expressions of these keratinocyte differentiation-involved genes in Module 1 were identified based on the data from TCGA.
239 DEGs were screened out in all 3 datasets, which were inclusive of 21 positively regulated genes and 218 negatively regulated genes. Based on these 239 DEGs, we finished constructing the PPI network which was formed from 225 nodes and 846 edges. We finished establishing 3 functional modules. And we analyzed 92 overlapping genes and 26 miRNA, including 11 upregulated genes targeted by 11 negatively regulated DEMs and 81 downregulated genes targeted by 15 positively regulated DEMs. As proof of the differential expression of metastasis-associated genes, eleven keratinocyte differentiation-involved genes, including LOR, EVPL, SPRR1A, FLG, SPRR1B, SPRR2B, TGM1, DSP, CSTA, CDSN, and IVL in Module 1, were obviously downregulated in metastatic melanoma tissue in comparison with primary melanoma tissue based on the data from TCGA.
239 melanoma metastasis-associated genes and 26 differentially expressed miRNA were identified in our study. The keratinocyte differentiation-involved genes may take part in melanoma metastasis, providing a latent molecular mechanism for this disease.
黑色素瘤是最致命的皮肤癌类型。迄今为止，其病理机制，尤其是转移机制，在很大程度上仍不清楚。我们对与黑色素瘤转移相关的基因的鉴定研究为黑色素瘤提供了新的认识。
从基因表达综合（GEO）数据库中，下载了基因表达微阵列数据集 GSE46517、GSE7553 和 GSE8401。我们使用 R 分析了转移性和非转移性黑色素瘤之间的差异表达基因（DEGs）。我们还使用 R 从 Li 的研究中的 GSE18509、GSE19387、GSE24996、GSE34460、GSE35579、GSE36236 和 GSE54492 数据集挖掘差异表达 miRNA（DEM）数据。基于 DEG 和 DEM 数据，我们通过应用 DAVID 数据库进行了功能富集分析。此外，我们还利用 STRING 数据库构建了蛋白质-蛋白质相互作用（PPI）网络并建立了功能模块。通过 Cytoscape 可视化 PPI 结果。我们通过应用 TargetScan、miRanda 和 PITA 数据库预测了 DEMs 的靶标，并确定了 DEGs 和预测靶标之间的重叠基因，然后构建了 DEM-DEG 对网络。根据 TCGA 中的数据，鉴定了模块 1 中涉及角质细胞分化的基因的表达。
在所有 3 个数据集中共筛选出 239 个 DEG，其中包括 21 个正调控基因和 218 个负调控基因。基于这 239 个 DEG，我们完成了 PPI 网络的构建，该网络由 225 个节点和 846 个边组成。我们完成了 3 个功能模块的构建。我们分析了 92 个重叠基因和 26 个 miRNA，包括 11 个受 11 个负调控 DEM 靶向的上调基因和 81 个受 15 个正调控 DEM 靶向的下调基因。作为转移相关基因差异表达的证明，基于 TCGA 中的数据，模块 1 中 11 个涉及角质细胞分化的基因，包括 LOR、EVPL、SPRR1A、FLG、SPRR1B、SPRR2B、TGM1、DSP、CSTA、CDSN 和 IVL，在转移性黑色素瘤组织中明显下调与原发性黑色素瘤组织相比。
本研究鉴定了 239 个黑色素瘤转移相关基因和 26 个差异表达 miRNA。涉及角质细胞分化的基因可能参与黑色素瘤转移，为该疾病提供了潜在的分子机制。
