Abstract
Dysregulated miRNA expression contributes to development of arsenic-induced cutaneous squamous cell carcinoma (cSCC). hsa-miR-186 (miR-186) is overexpressed in arsenical cSCC tissues as well as in preclinical cell line model of arsenical cSCC. Simultaneous miR-186 overexpression and chronic inorganic trivalent arsenite (iAs; 100 nM) exposure transformed human HaCaT cell line preferentially over miR-186 overexpression or iAs exposure alone. Both iAs and miR-186 regulate the expression of wide range of mRNA targets. However, how their interaction impacts the transcriptome-wide mRNA expression landscape ushering in cancer is unknown. We performed longitudinal RNA-seq analysis in passage-matched HaCaT cell clones (±miR-186 overexpression) with simultaneous chronic iAs exposure (0/100 nM) at 12 and 29 weeks. We determined the impact of each factor and their interaction towards differential gene expression and pathway dysregulation employing two different statistical approaches (t-statistic and 2-factor ANOVA). We show that a core set of pathways are dysregulated deterministically irrespective of the statistical approach chosen, possibly representing necessary changes for transformation. The data suggest that each clonal line could take a unique route to dysregulate this core set of pathways necessary for transformation, highlighting the possible role of stochasticity in cancer development. Evidence is presented to sift the strengths and weaknesses of each statistical methodology in providing biological understanding of events that play crucial roles in carcinogenesis in large datasets with multiple contributing variables.