Mechanism of regulation and suppression of melanoma invasiveness by novel retinoic acid receptor-gamma target gene carbohydrate sulfotransferase 10.

Retinoic acid (RA) induces growth arrest and differentiation of S91 murine melanoma cells and serves as a valuable model for this disease. RA acts through activation of RA receptors (RAR), which are members of the nuclear receptor superfamily of ligand-inducible transcription factors. Interestingly, differentiation is mediated by RARgamma, but not by RARalpha or RARbeta, suggesting that RARgamma possesses unique and uncharacterized molecular properties. To address this question, DNA microarrays in combination with RAR isoform-specific agonists were employed to identify novel RARgamma target genes that may play a role in this process. Here, we identified and validated carbohydrate sulfotransferase 10 (CHST10) as a novel RARgamma target gene in S91 cells. The RARgamma-inducible CHST10 promoter was obtained, and two atypical, independently functioning RA response elements were identified in a 425 bp region. Surprisingly, this fragment is bound by RARgamma, but not by RARalpha or RARbeta, thus providing a mechanism for the observed RARgamma-specific regulation. CHST10 is a sulfotransferase that forms HNK-1 glycan on neural cell adhesion proteins and glycolipids, and HNK-1 is thought to modulate cell adhesion and possibly metastasis. We show that CHST10 is also regulated by RARgamma in a significant subset of human melanoma cells, and three-dimensional cell culture migration assays suggest that CHST10 functions as a suppressor of invasiveness, but not proliferation, in these cells. Induction of CHST10 by RARgamma-activating retinoids may present a novel therapeutic strategy to inhibit invasiveness in a subset of melanoma patients.

Tumor necrosis factor receptor superfamily member TROY is a novel melanoma biomarker and potential therapeutic target.

Incidence of melanoma continues to rise, and a better understanding of its genetics will be critical to improve diagnosis and develop new treatments. Here, we search for novel melanoma-specific genes that may serve as biomarkers and therapeutic targets by using an in vitro genetic screen. One identified cDNA encoded TROY, a member of the tumor necrosis factor receptor superfamily (TNFRSF). TROY is widely expressed during embryogenesis, but in adults expression is restricted to hair follicles and brain. However, TROY had never been associated with melanoma, and it was selected for further study. First we show that expression in melanoma is specific by semiquantitative RT-PCR analysis of a large panel of established tumor cell lines. Next, specificity of expression was evaluated by immunohistochemistry analysis of primary cell cultures and patient tissues. TROY is expressed in 2/2 primary melanoma cells and 45/45 melanoma tissue samples (p < 0.0001). With the exception of sebaceous glands, TROY is not expressed in normal skin biopsies (p < 0.0001) or primary skin cell cultures that contain keratinocytes and epidermal melanocytes, nor is it expressed in other skin tumor cells (p < 0.0001). Finally, we show that TROY regulates melanoma growth, because replication of melanoma cells with reduced TROY levels through treatment with short-interfering RNA was significantly decreased relative to control cells (p < 0.004). In summary, TROY is the first TNFRSF member that is a biomarker for melanoma. TROY also presents a potentially novel cell surface signaling target for inhibitors, cell and/or antibody-based immunotherapies.