New oncogenic networks regulated by the RNA binding factor CUGBP1 in melanoma

Resumen

Melanoma is a prime example of an aggressive tumor that accumulates a plethora of changes in the transcriptome and the proteome. Consequently, distinguishing drivers from inconsequential passenger events has been a main challenge in this disease. Consequently, and despite great progress in identifying (epi)genetic defects accumulated during melanoma progression, the molecular bases underlying the aggressive behavior of this tumor type are not completely understood. RNA binding proteins remain largely unexplored in melanoma. We considered this lack of information of prime relevance as transcripts of nearly all known oncogenes and tumor suppressors may be regulated by alternative splicing and/or controlled by various mechanisms that define mRNA stability and ultimately, competency for translation. Here we show that a screen for all reported mRNA binding proteins (mRBPs) suggested that these genes are not prime targets of mutation or copy number variation. Instead, a customized microarray revealed a series of RBPs overexpressed in melanoma cells when compared to normal melanocytes. Of those factors, we found particularly attractive the CUGBP1/CELF1 protein. First, CUGBP1 is a multifunctional mRBP with a broad spectrum of functions, ranging from the modulation of alternative splicing to modulation of mRNA decay. Secondly, there is little information on comprehensive genome-wide analyses for CUGBP1 in cancer cells, with no previous report in skin cancer. Therefore this PhD thesis was set to address the following unknown aspects of CUGBP1 in melanoma: (i) expression, (ii) functional requirement, and (iii) mechanism of action. In brief, CUGBP1 was found overexpressed in human melanoma cells and tissue specimens. Targeted gene depletion demonstrated that CUGBP1 is required to sustain melanoma cell proliferation. Mechanistically, genome wide human junction arrays, RNA immunoprecipitation followed by sequencing and iTRAQ-MS/MS proteomics assays were utilized and identified novel and direct targets of CUGBP1. Together, these high throughput approaches (to our knowledge the first in class for this gene) uncovered a coordinated network of tumor-associated cell cycle regulators and chromatin remodelers as CUGBP1 targets. Central in this CUGBP1 regulated networks was the oncogene DEK, a feature we validated by targeted gene depletion, cDNA arrays and histological validation in clinical datasets. Mechanistically, CUGBP1 was found to bind to the 3’UTR of DEK stabilizing its mRNA levels and ultimately allowing for an efficient cell proliferation. These results illustrate the power of comprehensive analyses of RNA regulators in the identification of novel malignant features of cancer cells.