Blood glutamate scavengers increase pro-apoptotic signaling and reduce metastatic melanoma growth in-vivo

Inhibition of extracellular glutamate (Glu) release decreases proliferation and invasion, induces apoptosis, and inhibits melanoma metastatic abilities. Previous studies have shown that Blood-glutamate scavenging (BGS), a novel treatment approach, has been found to be beneficial in attenuating glioblastoma progression by reducing brain Glu levels. Therefore, in this study we evaluated the ability of BGS treatment to inhibit brain metastatic melanoma progression in-vivo. RET melanoma cells were implanted in C56BL/6J mice to induce brain melanoma tumors followed by treatment with BGS or vehicle administered for fourteen days. Bioluminescent imaging was conducted to evaluate tumor growth, and plasma/CSF Glu levels were monitored throughout. Immunofluorescence staining of Ki67 and 53BP1 was used to analyze tumor cell proliferation and DNA double-strand breaks. In addition, we analyzed CD8, CD68, CD206, p-STAT1 and iNOS expression to evaluate alterations in tumor micro-environment and anti-tumor immune response due to treatment. Our results show that BGS treatment reduces CSF Glu concentration and consequently melanoma growth in-vivo by decreasing tumor cell proliferation and increasing pro-apoptotic signaling in C56BL/6J mice. Furthermore, BGS treatment supported CD8+ cell recruitment and CD68+ macrophage invasion. These findings suggest that BGS can be of potential therapeutic relevance in the treatment of metastatic melanoma.

Melanoma cell culture. The cells studied included three melanoma cell lines: HT144 human metastatic melanoma cell line, Mel 624, metastatic melanoma derivative cells from a patient at the surgical branch of the National Institutes of Health (NIH), and the RET mouse melanoma cell line. These cell lines were chosen based on their mGluR1 expression and metastatic origin. In addition, we chose the RET cell line since it has been shown that the RET oncogene is commonly mutated in human melanomas, particularly in desmoplastic melanomas which have an increased risk for brain metastasis. The human metastatic HT144 cell line was purchased from the American Type Culture Collection (ATCC), and the RET cells were kindly donated by Dr. Carmit Levi, Tel-Aviv University. Human astrocytes (ScienCell Research Laboratories (Cat No 1800)) served as a control for basal mGluR1 expression levels in intact brain cells compared to the human metastatic melanoma derivative cells and RET melanoma cell line. The RET cells were double labeled with mCherry and Luc2 (pLNT/Sffv-MCS/ ccdB plasmid), sorted for red fluorescence, and selected in culture before intracranial (i.c.) injection as detailed below. RET melanoma cells (5 × 10 3 /2 μl) were implanted in C56BL/6J mice to induce tumors along with their accompanying neovascularization. The cells were cultured in RPMI-1640 supplemented with 10% FBS, Pen-Strep, and 4 mg ml −1 glucose at 37 °C under a humidified atmosphere of 5% CO 2 -95% air. Before injection, confluent monolayers of cells were released from the tissue culture flask using 0.25% trypsin, rinsed twice with serum-free RPMI-1640, centrifuged, and washed with sterile BPS.
Intracranial inoculation of RET cells and BGS treatment protocol. C57BL/6 mice were anesthetized with xylazine (10 mg/kg)/ketamine (70 mg/kg). After disinfection and incision of the skin, 2 μl of RET tumor cells (5 × 10 3 /2 μl) were stereotaxically implanted at the following location: 0.5 mm forward from the bregma, 2.1 mm lateral, and 3.0 mm ventral from the dura as we previously described 38 . RET-implanted mice were randomly divided into two groups following tumor cell inoculation.
Group 1 was the control group. Beginning at 1 h post tumor implantation, subcutaneous administration of 200 μl saline was performed daily. The mice were supplied with tap water for drinking containing 0.3 M NaCl until the termination of the experiment.
Glutamate plasma and CSF determination assay. We sampled CSF in order to evaluate the impact of tumor growth on brain tissue Glu levels and of rGOT1 on excess Glu. For this experiment, we sampled CSF and blood 3 and 7 days after melanoma cell implantation in BGS and vehicle-control treated mice. The sham control group was added at the 3-day time-point. Animals were anesthetized with a mixture of ketamine/xylazine (ketamine 100 mg/kg, xylazine 10 mg/kg) and CSF samples were extracted from the cisterna magna. Briefly, following mice anesthetization, each mouse's head was secured with a stereotactic head adaptor, the cisterna magna was exposed, and the mouse's body was positioned down so that the head formed a nearly 135° angle with the body. CSF was immediately collected from the cisterna magna using a glass capillary tube 37 and the samples were then frozen on dry ice and stored at − 80 °C [sham (n = 5); vehicle-control (n = 5); rGOT1/OxAc treatment (n = 5)]. Glu levels were analyzed with pre-column derivatization with OPA reagent and separated by reversedphase HPLC with a scanning fluorescence detector. The excitation and emission wavelengths used were 350 and 460 nm, respectively. Chromatography was performed using the UltiMate 3000 LC system (Thermo Scientific). The amino acid standard mix (Sigma-Aldrich) and samples were precolumn derivatized with OPA reagent solution. The derivatization reagent was 5.0 mg OPA dissolved in 100 µL of methanol and diluted with 900 µL of 0.4 M borate buffer (pH 9.5) and 5 µL of b-mercaptoethanol. Derivatization of amino acids: the standard or CSF sample was mixed with 5 µl of OPA derivatization reagent in the auto-sampler. All of the chromatographic separations were performed on a Thermo Hypersyl Gold 5U column (4.6 mm × 250 mm, 2.5 µl). The amino acid concentration was determined using the peak area and the external standard method.
Lysate preparation and immunoblot analysis. Cells were plated on a 10 cm culture dish at 1 × 10 6 cells in RPMI-1640 supplemented with 10% FBS, Pen-Strep, and 4 mg ml −1 glucose for 24 h before being lysed. For immunoblot analysis, equal amounts of protein from each sample were loaded and resolved by SDS-polyacrylamide gel electrophoresis through 7.5-10% gels as we have previously described 39 . The gels were electrophoretically transferred to a nitrocellulose membrane. Membranes were blocked, blotted with rabbit anti-mGluR1 (1:1000; Cell Signaling) and mouse anti-actin (1:10,000; MP Biomedical) primary antibodies followed by a secondary antibody linked to horseradish peroxidase. Immunoreactive bands were detected by a chemiluminescence reaction, and protein levels were quantified by densitometric analysis of protein bands using the ImageJ software.
For protein level analysis in-vivo, control/BGS treated mice (n = 5 animals for each group) were perfused with ice-cold PBS 7 days after tumor cells implantation, and brains were removed. Only the tumor area with some surrounding brain tissue was homogenized by a manual homogenizer (Fisher Scientific Med 11-850-56, USA) in a lysis buffer. For immunoblot analysis, equal amounts of protein from each sample were loaded and resolved by SDS-polyacrylamide gel electrophoresis through 7.5-10% gels as described above. The protein levels were quantified by densitometry analysis of protein bands using the ImageJ software. p-STAT1 was compared to total STAT1 of stripped membrane, and Iba1 was compared to actin levels of stripped membrane. Primary antibodies were rabbit anti-p-STAT1 (Ser727) (1:1000; Abcam), mouse anti STAT1 (1:1000, abcam), rabbit anti-Iba1 (1:1000; abcam), and mouse anti-actin (1:10,000; MP Biomedical).

Flow cytometry and FACS sorting. Mononuclear cells were isolated from the brains as previously
reported [Mayo 2014]. Mice were euthanized and subjected to perfusion through the left ventricle with ice-cold sterile PBS. Brains were then removed, minced by gentle MACS™ Dissociator (Miltenyi Biotec), and enzymatically dissociated with 0.05% (w/v) collagenase type III, 0.5% Dispase II, 40 mg/ml DNAse I, 20 mM HEPES in HBSS for 30 min at 37 °C to make a suspension of single cells. Enzymes were inactivated with 20 ml of Ca 2+ / Mg 2+ -free HBSS containing 2 mM EDTA. The digested tissue was triturated and passed through a 100-µM cell strainer. Cells were centrifuged and resuspended in 25% isotonic Percoll, underlined by 75% isotonic Percoll, and centrifuged at 1000×g at 20 °C for 30 min. Cells were collected from the 75%-25% interphase, and sorted by FACS (SONY SH800) Isolated cells were stained with fluorophore-conjugated antibodies for CD8 (53-6.7, Biolegend) and CD3 (17A2, Biolegend) for 45 min at 4 °C in the dark.
Statistical analysis. The Shapiro-Wilk test was applied to examine the distribution of the variables, and our results indicated a normal distribution of all variables except for tumor growth at Day 14. Accordingly, comparisons of the normally distributed variables were conducted using unpaired two-tailed Student's t-tests using GraphPad Prism 4 software and Mann-Whitney U tests using SPSS software version 25 (*P < 0.05; **P < 0.01; and ***P < 0.001). Results are presented as mean ± SD. The tumor volume at Day 14 in the treated vs. untreated group is presented as the median ± interquartile range.

Results mGluR1 expression in murine and humane metastatic melanoma cells. Previous studies have
demonstrated a significant inhibition in melanoma cell growth in vitro and in vivo as a result of blocking the mGluR1 receptor. We first examined mGluR1 expression in various human melanoma cell lines and compared them with the RET mouse melanoma cell line used in this study. The human astrocytes line served as a control for the basal expression of mGlur1 in the brain. Our results demonstrated a significantly higher expression of mGluR1 in HT144 and RET melanoma cell lines, and in one patient-derived melanoma cells compared to its expression in the human astrocyte cell line (Fig. 1A). The highest expression of mGluR1 was detected in RET cells. Since the RET murine melanoma cell line highly expresses mGluR1, this line was utilized for the in vivo experiments.

BGS treatment decreased glutamate levels in the plasma and CSF in melanoma-bearing mice.
To confirm whether BGS would reduce the levels of extracellular Glu due to RET melanoma cell secretion, we examined CSF and plasma Glu concentrations in treated vs. untreated mice. Tumor growth was confirmed by bioluminescence prior to CSF and blood sampling. Our results demonstrate that plasma Glu levels were slightly and non-significantly elevated at Days 3 and 7 in melanoma-implanted mice, compared to naïve mice (Fig. 1B). Furthermore, we verified a significant reduction in plasma Glu levels on Days 3 and 7 post melanoma implantation in the BGS treated group (Fig. 1B), compared to the naive mice or to vehicle-treated mice implanted with RET melanoma cells (naïve mice: 19.98 ± 0.9; vehicle-control group: day 3 23.25 ± 1.5, day 7 25.62 ± 2.5; BGS group: day 3 12.71 ± 1.3 μM, day 7 11.55 ± 1.6 μM, p˂0.001).
In the vehicle-treated mice, CSF Glu levels were significantly elevated at Days 3 and 7 post melanoma cells implantation compared to the naïve mice (Fig. 1C). Moreover, Glu concentration was elevated more than fivefold at Day 7 compared to Day 3 post-implantation (3 days, 3.6 ± 0.87; 7 days, 20.02 ± 5.6, p˂0.001). Three days of treatment with BGS resulted in a non-significant decrease in CSF Glu levels, however, CSF Glu concentration was significantly decreased compared to the vehicle-control group at Day 7 (3 days 2.4 ± 0.88; 7 days 6.9 ± 2.1, p < 0.001) (Fig. 1C). These results verified the ability of BGS treatment to reduce excess levels of Glu that accumulate in the brain parenchyma due to tumor growth.

Blood glutamate scavengers reduced melanoma growth in vivo.
In order to examine whether BGS treatment, which can potentially shift Glu secreted by tumor cells from the brain to the blood, affects tumor growth, RET melanoma cells were injected into the right hemisphere. BGS or vehicle treatment began one hour and tumor growth was examined on Days 2, 7 and 14 following tumor implantation. Our results demonstrate that despite the lack of a significant difference between tumor volume in treated vs. untreated mice on Days 2 and 7 of treatment, tumor size was significantly decreased as measured by bioluminescence imaging (control 25.5 × 10 4 , BGS 10.1 × 10 4 p < 0.01) ( Fig. 2A,B). This significant reduction in tumor size in treated animals suggests that the depletion of excess brain Glu by boosting the Glu efflux from the CSF to the blood was sufficient to inhibit melanoma cell growth. On Day 14, there were still differences in tumor size between the groups, however they were not statistically significant (check 43.2 × 105, BGS 19.5 × 105 p = 0.06) (Fig. 2C).
Next, we examined the expression of various proliferation, apoptotic, and inflammatory markers within and around tumor cells in the brain in order to study the effects of BGS treatment at the cellular level. All immunohistochemistry was performed on sectioned brains at Day 7 after implantation and treatments. www.nature.com/scientificreports/   (Fig. 3). These results suggest that one of the effects of BGS treatment is reduced tumor growth due to inhibition of RET cell proliferation.
BGS increased the expression of a DNA suppressor marker and pro-apoptotic signaling. Next, we examined various apoptotic markers in order to determine whether BGS, besides for decreasing the tumor proliferation rate, can also promote apoptotic cell death. 53BP1 is a p53 binding protein that binds to the central DNA-binding domain of p53 42 . In response to DNA damage, the protein deposits at the sites of DNA strand breaks and plays a central role in the response of mammalian cells to genotoxic stress 43 . 53BP1-deficient mice are predisposed to tumor development as 53BP1 is essential for an appropriate cellular response to DNA damage in vivo 44 . Therefore, we examined 53BPI expression within tumors in both BGS-treated and control groups. Our results demonstrated a significantly higher expression of 53BPI in the RET cells after 1 week of BGS treatment compared to vehicle-control animals, as measured by fluorescence density (control 48.3 ± 10.3; BGS 58.6 ± 12.5, p < 0.01) (Fig. 4).  www.nature.com/scientificreports/ It has been previously demonstrated that DNA strand breaks lead to activation of the apoptotic process, which in turn is regulated by p53 activation and increased caspase-3 expression 45 . The expression of caspase-3 in gastric cancer patients, for example, correlates with better prognosis, and therefore may act as a tumor suppressor 46 . Our examination of active caspase-3 expression revealed that BGS treatment significantly increases caspase-3 expression in RET melanoma cells in vivo (control 42.0 ± 2.89; BGS 52.0 ± 9.6) (Fig. 5). This suggests that BGS treatment facilitates cell death in melanoma cells as a response to DNA damage by depleting brain extracellular Glu.

BGS treatment mediated CD8 + cell recruitment into the tumor and induced p-STAT1 signaling.
Tumor-infiltrating immune cells are associated with tumor prognosis, and a high density of CD8 + T cells is associated with a good clinical outcome in various tumor types [47][48][49] . Since BGS treatment mediated RET  www.nature.com/scientificreports/ melanoma cell death, we further examined whether tumor cell death was accompanied by an invasion of CD8 + cytotoxic cells into the tumor area. As demonstrated by both immunohistochemistry and flow cytometry, BGS treatment increased the infiltration of CD8 + cells into the tumor (Fig. 6A-C). Flow cytometry analysis revealed 37.4 ± 3.01% of CD8 + cells infiltration in BGS treatment, compared to only 18.6 ± 2.65% in the control group (Fig. 6D). The immune response consists of immune cell infiltration into tumor tissue and inhibitory signaling of tumor progression, depending on the tumor microenvironment [50][51][52] . Therefore, we next assessed the expression of a phosphorylated signal transducer and activator of transcription 1 (p-STAT1), one of the IFN-gamma signaling pathway molecules. Our results demonstrated that the density of p-STAT1 within tumor cells was significantly elevated in the BGS treatment group, when compared with the control group (control 31.13 ± 3.9; BGS 55.12 ± 8.4) (Fig. 7A-C). An increase in p-STAT1 in BGS-treated animals was also confirmed by Western blot analysis (Fig. 7E), with a 27% increase compared to control. Furthermore, RET melanoma cells surrounded by CD8 + cells expressed a high level of p-STAT1, which was significantly increased within melanoma cells in the  www.nature.com/scientificreports/ BGS-treated group (Fig. 7D). This strongly supports the theory that BGS treatment mediates CD8 + cell infiltration into the tumor and is associated with signaling induction of p-STAT1. This also coincides with signs of tissue damage, such as cell apoptosis.

BGS treatment mediates phagocytic CD68 + macrophages invasion into the tumour. We used
CD68 as a marker of phagocytic activity of macrophages. An increased invasion of macrophages into the tumor area was observed in the BGS treated mice (Fig. 8A,B). Moreover, RET melanoma cells were detected inside macrophages, as was demonstrated by co-localization of CD68 macrophage and mCherry RET cells with fragmented condensed nuclei (Fig. 8C). These macrophages were co-localized with iNOS immunostaining, suggesting that they are of the pro-inflammatory type of macrophage. In the control group, iNOS density in the tumor area was significantly lower compared to the BGS treatment group (control 33.6 ± 9.4; BGS 67.1 ± 8.8), and many of the CD68 cells were iNOS negative ( Fig. 8D; white arrows). This suggests that BGS treatment mediates a proinflammatory macrophage phenotype that contributes to the tumor suppression.
To further test this theory, we examined the percentage of cells positive for CD206, a marker of the M2 antiinflammatory macrophages/microglia type, from the total cells expressing Iba1. Our results demonstrated that the total Iba1 immunofluorescence density was not different in the control vs. BGS groups (control 38.4 ± 9.3; BGS 29.8 ± 14.6), and this was also confirmed by Western blot analysis (Fig. 8G). However, quantitation of CD206 + fluorescence out of total Iba1 demonstrated that BGS treatment significantly decreased CD206 + cells to 32% compared to 81% in control treatment (Fig. 8E,F). These results strongly suggest that BGS treatment enhances the pro-inflammatory response by the macrophages that invade and surround the tumor.

Discussion
The blood-glutamate scavenging approach was recently found to be beneficial in attenuating the progression of glioblastomas by reducing excess brain Glu concentrations 5 . Supplementing rGOT1 and its co-substrate, oxaloacetic acid, causes a rapid decrease in blood levels of Glu 36 . In turn, this leads to a larger Glu gradient between the brain and the blood, promoting an efflux of the excess extracellular Glu into the blood 35 . In this study, we examined the anti-tumor effects of such a BGS treatment in a brain metastatic melanoma mice model. The main findings of this study were: (1) Glu concentrations in the CSF consistently increased along with tumor growth, and was significantly higher in implanted animals as compared to healthy mice; (2) The BGS treatment significantly decreased plasma and CSF Glu concentrations; (3) The BGS treatment reduced melanoma growth in vivo by decreasing tumor cell proliferation and by increasing pro-apoptotic signaling; (4) The BGS treatment mediated CD8 + T cell recruitment into the tumor; and (5) The BGS treatment enhanced the pro-inflammatory response by decreasing CD206 positive cells, and mediated phagocytic CD68 + macrophage invasion into the tumor.
Our results support previous reports that showed an increase in extracellular brain Glu levels in progressed tumors in patients 53,54 . Cancer cells reorganize the body's metabolism in order to maintain their high proliferation rates. Along with glycolysis and glutaminolysis, amino acids such as Glu are used as an alternative energy source in order to support the biosynthesis of fatty acids. These fatty acids represent part of the secretome of transformed cells, including melanomas 55,56 . Downregulation or blocking of mGluR1 receptors has been shown to result in a decreased rate of metastatic melanoma growth in mice 8,10,12,57,58 . The melanoma lines that were examined in this study showed increased mGluR1 expression, and may have contributed to the high proliferative rate of these cells 11 . Elevation in Glu levels that surround brain tumors may occur due to the increased Glu release by the tumor cells, and/or by neuronal cell death. Decreased Glu uptake caused by an alteration in the expression and function of glutamate excitatory amino acid transporter 2 (EAAT-2) in astrocytes and tumor cells may also contribute to this elevation [59][60][61][62] . Tumor cells release Glu mainly via the xc− antiporter, a transporter that plays a critical role in maintaining intracellular and extracellular antioxidant levels (e.g., glutathione, cysteine) by exchanging extracellular cystine for intracellular glutamate 6,63 . Excess extracellular Glu has been shown to enhance glioma and melanoma tumor cell growth and invasion 1,[63][64][65][66] , and studies have demonstrated that inhibiting the release of Glu results in a decrease of melanoma tumor growth in mGlu1 receptor-expressing melanomas 63,67 . Similarly, our results demonstrate that reducing extracellular Glu levels following BGS treatment inhibits tumor cell proliferation and induces apoptotic tumor cell death. This was manifested by a significant decrease in tumor size and a high expression of apoptotic markers. Increased tumor cell proliferation is a hallmark of malignancy, and Ki67 has been the most widely used proliferation marker in many types of tumors, including melanomas 40,41 . Our results demonstrate that Ki67 was significantly decreased within the tumor area following BGS treatment, demonstrating a potential clinical significance of this treatment. Although BGS-treated animal tumor volume was not significantly smaller at a more advanced stage (14 days after cell transplantation), this trend was still maintained. These results support our previous study in glioblastomas, in which we demonstrated a weaker inhibition of tumor growth by BGS treatment alone vs. a combined treatment approach of BGS with temozolomide 38 , indicating that BGS treatment may be more potent if given in combination with cytotoxic or immunotherapy treatments.
It has been reported that Glu at low physiological concentrations activates human T cells, adhesion, migration, proliferation, intracellular Ca 2 and more via its GluRs 30,31 . Yet, Glu's direct effects on T cells depend on its concentration, and may even be inhibitory at excess pathological concentrations 29,31 . The effects of Glu on T cells also depend on the specific GluR types expressed on the target T cell, the T cell's type and subtype, the T cell's resting or activated state, and the presence or absence of other concurrent stimuli besides for Glu 68 . Interestingly, while Glu has been shown to suppress the proliferation of activated T cells, it does not affect the proliferation of normal naive/resting human T cells 69,70 . In this study, increased invasion of CD8 + cells was observed within the tumor area in BGS treated mice. In addition, melanoma cells surrounded by CD8 + cells displayed augmented www.nature.com/scientificreports/ p-STAT1 signaling, which may be due to IFN-γ secretion by CD8 + cells within the tumor microenvironment [71][72][73] . The tumor-suppressive role of STAT1 has been previously demonstrated, whereby reconstitution of STAT1 in STAT1-deficient murine fibrosarcoma cells significantly suppressed tumorigenicity and metastasis in nude mice 74 . Moreover, a high (vs. low) expression of STAT1 is reported to be associated with a good prognosis in melanoma patients 75 . The binding of IFN-γ to its related receptor leads to the phosphorylation-dependent activation of JAK1/2 and STAT1. p-STAT1 then forms homodimers that translocate to the nucleus and activate the transcription of IFN-γ-stimulated genes 71 . IFN-induced STAT1 can subsequently activate chemokines, such as CXCL9, CXCL10, and CXCL11, which can recruit more CD8 + T cells to provide anti-tumor immunity 75,76 . Moreover, STAT1 as a transcription factor induces the transcription of cell cycle and apoptosis-related genes. For instance, STAT1 can arrest the cell cycle in response to IFNγ through direct interactions with cyclin D1 and CDK4 proteins 77 . IFNγ/STAT1-induced expression of caspases 1 and 8 has been reported in cancer cell lines, where STAT1 regulates Bcl-2 family members, including Bcl-xl, Bax, and Bak [78][79][80] . It has been demonstrated that patients with upregulated CD8 T effector and IFN-γ gene signatures may benefit from immunotherapy 81 . In addition, it was proposed that a high density of tumor associated macrophages (TAMs) correlates with brain tumor grade, suggesting a supportive role of TAMs in tumor progression 82,83 . Furthermore, it was reported that the M1 macrophage subtype has an inhibitory effect on tumor growth, whereas the M2 subtype plays a supportive role in promoting tumor progression by secreting cell growth cytokines and survival factors [84][85][86] . In this study, BGS did not alter the total amount of TAM, however, the density of CD68 + cells at the tumor border increased in treated mice while the percentage of CD206 + from the total amount of macrophages decreased significantly. This indicates conversion of macrophage phenotype from M2 to M1. He et al. showed that the antibiotic doxycycline, reported as a potent anti-cancer treatment in many in vivo models, was able to inhibit macrophage polarization towards the M2 phenotype 87 . Consequently, our results demonstrated a potential role of rGOT1/OxAc treatment in altering the immune-microenvironment in brain metastatic growth, suggesting that it may be a promising new approach to metastatic cancer management 88 .
Not only does Glu facilitate the progression and invasion of tumor cells, but it also plays a main role in the promotion of tumor-associated seizures (TAS) 89 . Seizures occur in about 50-90% of glioma patients 90 . Increased peri-tumoral Glu concentration, as a result of reduced EAAT2 expression and increased system xc− expression has been shown to correlate with the presence of TAS in glioma patients 62,91 . Based on the above studies and the current results, we can assume that BGS treatment, administered shortly after the surgical removal or ablation of a tumor, or in the early stage of the primary or metastatic disease, has a potential to delay tumor recurrence and decrease the probability of TAS development.
Unfortunately, malignancy of the central nervous system has defied all current therapeutic approaches, and an improved understanding of the biology of brain tumors along with new therapeutic options are needed to improve patient survival and control disease progression. This is the first time that the BGS approach was tested in brain metastatic melanoma and it may be of high clinical significance for the future treatment of melanoma patients. However, the ability of the treatment significantly inhibits the tumor cells growing at the very early stage, was less effective at the more progressive stage of the tumor. Consequently, BGS may serve as a monotherapy at the early stage of the disease or as a complimentary treatment that potentially increases the effectiveness of other treatments such as immunotherapy and chemotherapy. It is clear that treatment of such a multifactorial disease as metastatic melanoma ought to involve a complex therapy. Therefore, future studies should focus on evaluating the efficacy of BGS treatment in combination with Temozolomide and check-point immunotherapy (Supplementary Information).