Pluto Bioinformatics

GSE149884: The transcriptome of a murine model of melanoma initiation and progression unravels molecular signatures of phenotype switch and novel independent prognostic factors for melanoma patients

Bulk RNA sequencing

Despite advances in therapeutics, metastatic melanoma is still associated with a poor prognosis. The study of changes in gene expression and affected molecular pathways during melanoma progression could contribute to a better understanding of the biology underlying its aggressiveness and lead to the identification of novel biomarkers for prognosis and response to treatment. Animal models that resemble both the initiation and progression of human melanoma are still rare. Here, we used a linear model of murine melanoma progression comprising an immortalized melanocyte lineage (melan-a), pre-malignant melanocytes (4C), a non-metastatic tumorigenic cell line (4C11-), and a metastatic cell line (4C11+),. We generated RNA-Seq data from each cell line and performed transcriptome analysis to identify changes in gene expression related to both early and late stages of melanoma progression. Principal Component and Hierarchical Clustering analyses grouped the cell types according to their morphology being epithelial-like (melan-a and 4C11+) or mesenchymal (4C and 4C11-), suggesting the existence of specific and dynamic gene expression programs associated to the transition through these characteristic phenotypes. Differential expression analysis revealed signatures of differentially expressed genes (DEGs) in each step of tumor progression, besides a set of 89 genes commonly perturbed throughout the melanoma progression and 332 genes consistently deregulated in cells with metastatic potential. We also identified in the initial stages of melanoma progression the significant deregulation (log2 fold-change 2, p < 0.01) of genes associated to Epithelial-to-Mesenchymal Transition (EMT), such as upregulation of Snai2, Twist1, Twist2, Zeb1 and downregulation of Mitf, Mlana, Tyrp1, Dct, Psen2, that were reversed in later stage of metastatic melanoma. These findings were confirmed by RT-qPCR or western blot. Genes overexpressed during the progression were enriched for molecular pathways relevant for cell proliferation and invasion and chemoresistance. Finally, DEGs were mapped to their human orthologs, which were subsequently used in univariate and multivariate survival analysis using gene expression and clinical data of 703 primary melanoma patients from the Leeds Melanoma Cohort, revealing several novel candidate prognostic markers for clinical use in melanoma patients. These findings validate the potential of this murine model to provide key insights into melanoma initiation and progression. SOURCE: Eduardo,Moraes,Reis (emreis@iq.usp.br) - University of São Paulo