Clinical Significance Melanoma is among the most aggressive skin cancers, with metastasis being the primary cause of cancer-related death. Despite therapeutic advances, the molecular mechanisms driving the transition from primary to metastatic melanoma remain incompletely understood, hindering development of effective treatments.
Study Design This study employed bioinformatics analysis of the publicly available GSE8401 gene expression dataset, which contains samples from 31 primary and 52 metastatic melanoma specimens. Differential gene expression analysis was performed to identify genes and pathways that distinguish metastatic from primary tumors.
Analytical Approach The analysis combined differential expression analysis, protein-protein interaction (PPI) network construction, hub gene identification, survival analysis, and functional enrichment to prioritize the most biologically and clinically relevant genes driving melanoma metastasis.
Experimental Validation Beyond bioinformatics, the study performed in vitro validation of key findings by knocking down the top hub gene KRT5 in melanoma cell lines to assess its functional role in migration and invasion, linking computational predictions to biological mechanism.
Dataset and Preprocessing The GSE8401 dataset from the Gene Expression Omnibus (GEO) contains expression profiles from 31 primary and 52 metastatic melanoma samples profiled on Affymetrix HG-U133Plus2 microarrays. Data preprocessing included normalization and background correction before statistical comparison.
Differentially Expressed Gene Identification A total of 425 differentially expressed genes (DEGs) were identified, with 201 upregulated and 224 downregulated in metastatic compared to primary melanoma. Statistical thresholds were applied to ensure the relevance of identified DEGs for downstream analysis.
PPI Network Construction DEGs were uploaded to the STRING database to construct a protein-protein interaction network. The resulting network was visualized and analyzed in Cytoscape using the MCODE and cytoHubba plugins to identify functional modules and rank hub genes by connectivity.
Cell Line Experiments KRT5 knockdown was performed in melanoma cell lines using siRNA. Migration was assessed by wound-healing (scratch) assay and invasion by Transwell assay with Matrigel coating. Results were quantified at 24-hour intervals to characterize the temporal effect of KRT5 loss on cell motility.
Ten Hub Genes Identified Network analysis identified 10 hub genes significantly associated with melanoma metastasis: CDK1, COL17A1, EGFR, DSG1, KRT14, FLG, CDH1, DSP, IVL, and KRT5. These genes showed the highest degree of connectivity in the PPI network and were consistently ranked highly by multiple hub gene scoring algorithms.
KEGG Pathway Analysis Functional enrichment analysis revealed that ECM-receptor interaction was the most significantly enriched pathway among downregulated DEGs in metastatic melanoma. Additional enriched pathways included PI3K-Akt signaling, focal adhesion, and pathways in cancer, all of which are known mediators of tumor invasion and metastasis.
GO Enrichment Gene Ontology analysis showed that upregulated DEGs were enriched for cell cycle-related processes including mitotic nuclear division and chromosome segregation, consistent with the increased proliferative activity of metastatic cells. Downregulated DEGs were enriched for keratinization and skin barrier functions.
Survival Analysis Analysis of TCGA melanoma data showed that expression levels of hub genes including KRT5, CDK1, and EGFR were significantly associated with overall survival. Patients with altered expression of these genes had worse clinical outcomes, supporting their prognostic relevance beyond the GSE8401 discovery dataset.
KRT5 Expression in Melanoma KRT5 (Keratin 5), a structural protein in epithelial cells, was significantly downregulated in metastatic compared to primary melanoma samples. This downregulation pattern suggested that loss of KRT5 may play a functional role in the acquisition of metastatic potential.
Wound Healing Assay Results siRNA-mediated knockdown of KRT5 in melanoma cell lines significantly accelerated wound closure compared to control cells. The enhanced migration was evident within 24 hours and became more pronounced at 48 hours, indicating that KRT5 loss promotes cell motility.
Invasion Assay Results Transwell invasion assays showed that KRT5 knockdown significantly increased the number of cells invading through Matrigel. This finding demonstrates that KRT5 does not merely affect cell migration, but also promotes the invasive capacity required for tissue penetration during metastasis.
Mechanistic Implications KRT5 is a component of the intermediate filament cytoskeleton that contributes to cell structural integrity and adhesion. Its downregulation may facilitate cytoskeletal remodeling that allows cells to adopt a more motile, mesenchymal-like phenotype consistent with epithelial-mesenchymal transition, a key process in cancer metastasis.
EGFR as a Therapeutic Target EGFR emerged as a hub gene upregulated in metastatic melanoma. While EGFR-targeted therapies are established in lung cancer, EGFR's role in melanoma metastasis suggests potential for repurposing existing targeted therapies. The identification of EGFR in this network context adds to growing evidence for EGFR inhibition in melanoma.
CDK1 and Cell Cycle Dysregulation CDK1 (Cyclin-Dependent Kinase 1) was among the most connected hub genes and is a master regulator of cell cycle progression. Its upregulation in metastatic melanoma is consistent with the increased proliferative capacity of metastatic cells and positions CDK1 as a potential therapeutic target using existing CDK inhibitors.
ECM Remodeling Pathway The enrichment of ECM-receptor interaction among downregulated DEGs suggests that loss of cell-ECM adhesion is a key step in enabling melanoma cells to detach from the primary tumor and invade surrounding tissues. Understanding which ECM components and integrins are affected may guide development of anti-metastatic strategies.
Multi-Gene Prognostic Panel The ten hub genes collectively form a candidate multi-gene prognostic signature for melanoma metastasis risk. Future clinical validation in prospective cohorts could enable development of a tissue-based test to identify primary melanoma patients at high risk of metastatic progression, informing surveillance intensity and adjuvant therapy decisions.
Dataset Limitations The study relied on a single publicly available microarray dataset (GSE8401), which has inherent limitations including batch effects and the absence of associated clinical outcome data for all samples. Validation in independent cohorts with comprehensive clinical annotation is needed to confirm the prognostic value of identified hub genes.
Functional Validation Scope Only KRT5 was functionally validated in vitro among the ten hub genes. Systematic functional characterization of the remaining nine hub genes - including CDK1, EGFR, COL17A1, and CDH1 - through knockdown, overexpression, and rescue experiments would strengthen the biological interpretation of network analysis findings.
In Vivo Validation In vitro migration and invasion assays capture only certain aspects of the metastatic cascade. Future studies using xenograft mouse models with KRT5-knockdown or hub gene-perturbed melanoma cells would provide more physiologically relevant evidence for the functional roles identified computationally.
Mechanistic Pathways The molecular mechanisms linking hub gene expression changes to downstream metastatic signaling remain to be elucidated. Future work should investigate how KRT5 loss connects to ECM-receptor interactions, PI3K-Akt activation, and EMT transcription factors to build a comprehensive mechanistic model of hub gene-driven melanoma metastasis.