ABSTRACT: Glioblastoma (GBM) is the most prevalent primary malignant brain tumor in adults and remains almost invariably lethal due to its aggressive and invasive behavior. Despite many clinical trials have been put forward, the standard treatment that stands for more than a decade consists of surgical resection, followed by temozolomide (TMZ) chemotherapy and radiation treatment. However, this strategy has improved the average survival time to 14.6 months when compared to 12,1 months only with surgery and radiation1 . Therefore, due to the low prognosis, different efforts have been employed in the search for new drugs and therapeutic protocols for this malignancy. A class of drugs that has emerged as a promising candidate for GBM is Histone Deacetylase inhibitors (iHDAC). Indeed, pre-clinical studies have demonstrated the efficiency of different iHDACs as anti-tumor agents, especially when associated with other therapies, including chemotherapy and radiation2. In fact, the use of iHDACs reduced the expression of genes involved in DNA repair, formation of mitotic spindles and homologous chromosome segregation, leading to glioma apoptosis in vitro 3. In addition, it has been reported that small iHDACs molecules, such as Trichostatin (TSA), disrupts cellular signaling networks relevant to tumorigenesis, leading to an increased interest in testing iHDACs in clinical trials against different types of tumors 4 . In agreement, a high activity of HDAC has also been associated with tumor progression, corroborating the pharmacological inhibition of HDACs as an interesting anti-cancer molecular target. Previous works from our group are in line with this reasoning showing that iHDACs negatively regulate the migratory capacity of GBM in vitro, suggesting that these inhibitors may regulate the molecular signature of GBM secretome, making the tumor's microenvironment repulsive for cell migration5. In fact, the communication between tumor cells and their microenvironment may occur directly or indirectly through a variety of secreted molecules, such as growth factors, cytokines, chemokines and microRNAs6. One relevant secreted molecule described by our group is Nodal, an important morphogen involved in several cellular processes related to tumorigenesis. In this previous study, our group brings an original approach on the mechanisms that dynamically control intra and extracellular nodal availability during GBM tumorigenesis7. Thus, the heterotypic secretome that derives from the tumor cells, may be an important source of key regulators of the tumorigenic process6 . To better understand whether iHDACs can also affect the repertoire of secreted molecules, and therefore directly modulate the surrounding micro-environment displaying deleterious effects on tumor cells, we investigated the secretome of iHDAC treated GBM cells in vitro using a high throughput methodology of protein identification and quantification couple to label free mass spectrometry. Using this strategy we were able to successfully identify secreted molecules upon HDAC activity knockdown. Interestingly, we identified proteins exclusive to the iHDAC treated secretome as well as proteins found in both control and treated groups. These proteins were classified using the Matrisome classification system corroborating that the molecular signature of tumor secreted proteins, that belong to the ECM, was disrupted. In vitro experiments using 2D and 3D cell culture systems coupled to decellularized ECM analysis by confocal microscopy confirmed that HDAC activity knockdown gave rise to a disrupted pattern of ECM protein staining. Taken together our results first describe in the literature the relevance of epigenetic remodeling HDAC dependent for the molecular signature of GBM cells shedding light on putative molecular targets to approach the malignancy.
Joseph Evaristo 
