Abstract
Melanoma is a serious form of skin cancer derived from the pigment producing cells called melanocytes. New Zealand has the highest rate of melanoma in the world with as high as 50 cases of melanoma per 100,000 people, rendering melanoma as the fourth most common cancer. Melanoma can be cured if detected in the early stages, but the metastatic form of melanomas has a poor survival rate. Although advanced therapies have improved the patient survival rate, drug resistance is still a major problem in metastatic melanoma. It has become increasingly evident that the major obstacle in melanoma therapy is its heterogeneous nature. Metastatic tumour cells in melanoma are known to switch back and forth from a non-invasive to an invasive state, due to tumour microenvironmental cues, such as hypoxia. Consequently, it is difficult to target the subpopulations of invasive cells that escape therapies by switching back to a non-invasive state, and then later initiate the invasive properties by switching back to the invasive state. Therefore, a better understanding of the underlying mechanisms that initiate this phenotype switch is required. Previous studies carried out by our group and other investigators have demonstrated two main phenotypes in melanoma that were distinct in their gene expression profiles, and which were independent of the tumour-associated genetic mutations from which these cell lines were derived. The melanoma cells can switch back and forth from one phenotype to the other, with the non-invasive phenotype demonstrating a low invasive capacity, and invasive phenotype demonstrating a high invasive capacity. Given that these phenotypes are reversible and independent of the mutational status, it is plausible that an epigenetic mechanism is involved in the regulation of these phenotypes. Epigenetic regulation is defined as the mechanism that transduces the inheritance of gene expression without altering the underlying DNA sequence. DNA methylation is an epigenetic mechanism which involves addition of a methyl group to the cytosine base. In somatic tissues, the DNA methylation profile underlies cellular identity which is conserved over a lifetime. Changes in the environment are capable of altering the DNA methylation which consequently leads to changes in the phenotype. I hypothesized that DNA methylation, which is the most stable and reversible epigenetic mechanism, would underlie the mechanism that distinguishes the invasive from non-invasive melanomas.
To address this hypothesis, a genome-wide DNA methylation analysis was carried out using reduced representation bisulphite sequencing (RRBS). Differential methylation analysis revealed 39 differentially methylated fragments (DMFs) in the invasive cell lines as compared to the non-invasive cell lines, with 20 DMFs hypermethylated and 19 DMFs hypomethylated in the invasive melanomas. This observation confirmed that DNA methylation profiles can distinguish the invasive group from the non-invasive group of melanoma cell lines. To identify the relationship between DNA methylation and gene expression, genome-wide RNA sequencing (RNA-Seq) was carried out on the same cell lines. The RNA-Seq results identified interesting genes which were associated with growth factor binding, transmembrane receptor kinase activity, cell adhesion and extra cellular matrix organization. Comparison of the DMFs identified from multiple RRBS analyses with differentially expressed genes identified by multiple RNA-Seq analyses resulted in a total of 17 common differentially expressed genes associated with DNA methylation. Of these 17 genes, AVPI1 was selected for further validation because it had significant promoter hypermethylation in the invasive cell lines compared to non-invasive cell lines. In addition, low expression of this gene was identified in the invasive cell lines, and this was associated with MAPK activity and epithelial sodium channel (ENaC) downregulation. Functional analysis suggested that AVPI1 promoter methylation may be associated indirectly with melanoma invasion through an intermediate mechanism of ENaC regulation and epithelial mesenchymal transition (EMT).
Overall, this study has provided novel insights into distinct DNA methylation profiles between invasive and non-invasive melanoma cell lines, which predominantly included cell lines derived from New Zealand metastatic melanoma patients. Integrative analysis of DNA methylation and gene expression revealed potential genes involved in the invasive phenotype of melanoma, which was followed by functional analysis of AVPI1 to investigate its role in the invasive phenotype. These investigations of AVPI1 and other genes identified from the integrative analysis of DNA methylation and gene expression have provided a platform, which can further be used to establish their role in phenotype switching.