Decreased expression of ATF-3 in melanoma promotes tumor growth via ERK and AKT pathways

Abstract

<h3>Introduction</h3> Activating transcription factor 3 (ATF-3), a cyclic AMP-dependent transcription factor that is encoded by the ATF-3 gene, plays a regulatory role in melanoma, although its function during tumor progression is unknown. Here we evaluated ATF-3 expression during melanoma evolution to elucidate its role in melanogenesis. <h3>Materials and Methods</h3> Primary and metastatic melanoma cell lines, patient melanomas and murine melanoma xenografts were evaluated by immunofluorescence, Western Blot and RT-PCR studies. Melanoma cell lines were retrovirally transfected with ATF-3 overexpressing vectors and CCK8, wound healing and Matrigel invasion studies were performed to test for proliferation, migration and invasive properties, respectively. Xenograft murine models were designed by subcutaneously injecting melanoma cells or its overexpressed counterpart into the anterior and posterior flanks of nude mice to evaluate tumor size and rate of formation. <h3>Results</h3> By immunofluorescence, ATF-3 staining declined in metastatic versus primary melanoma cell lines and with progression from patient nevi to primary to metastatic melanomas. Analysis of TCGA skin cutaneous melanoma database showed that lower ATF3 expression correlated with poor prognosis. Metastatic melanoma cell lines with retrovirally-induced ATF-3 overexpression (A2058 and C8161 ATF-3 OE) exhibited decreased prolif- eration, migration and invasion. GO and KEGG analyses of our RNA seq data showed downregulation of phospho- rylated ERK and AKT in A2058 ATF-3 OE cells that was further validated by Western Blot. In vivo, xenografted A2058 ATF-3 OE cells formed smaller and less abundant tumors in murine subcutis than did control cells, and Ki-67 staining confirmed lower labeling indices in ATF-3 overexpressing lesions. <h3>Conclusions</h3> ATF-3 expression is associated with melanoma virulence and thus provides a potential target for novel therapies and predictive biomarker applications.