Dabrafenib (HY-14660) and L-ascorbic acid 2-phosphate (p-ASC) (HY-103701) were obtained from MedChemExpress, NJ, USA. L-ascorbic acid (ASC) (A4403), formalin solution neutral buffered (HT5012), Methyl cellulose (M0512), N-acetylcysteine (NAC) (A7250), and Trolox (648471) were obtained from Sigma-Aldrich, St. Louis, MO, USA. Stock solutions of ASC, p-ascorbate, Trolox, and NAC were prepared in phosphate-buffered saline (PBS). The 0.4% Trypan Blue solution (17-942E) was from Lonza, Basel, Switzerland. 2′,7′-Dichlorofluorescein diacetate (DCFDA; D0063) was from Chemodex, St. Gallen, Switzerland. Goat anti-mouse (172-1011) and anti-rabbit (172-1019) IgG (H + L)-horseradish peroxidase-conjugated, Tris-based (1610716) and sodium dodecyl sulfate (SDS) (161-0300) were from BioRad Laboratories, Hercules, CA, USA. JC-1 (420200) is from Calbiochem, San Diego, CA, USA. Dabrafenib was dissolved in dimethyl sulfoxide (DMSO) to prepare a 10 mM stock solution, which was stored at -20 °C.

The human melanoma cell line A375 was obtained from the European Collection of Cell Cultures (catalog n. 88113005, Salisbury, UK), whereas the human melanoma cell line M14 was kindly provided by Dr. Zupi G. “Istituto Regina Elena” (Rome, Italy). Chemoresistant A375-R and M14-R were generated in our laboratories by exposing the parental cell line to increasing concentrations of dabrafenib (up to 1 µmol/L) for 4 months and validated according to our previous works^[15,16]. All cells were authenticated via short tandem repeat (STR) profiling (BMR Genomics, Padova, Italy). The results exhibited an allele percent match (EV value) of 89%-100% against the reference database. Consequently, in accordance with the International Cell Line Authentication Committee (ICLAC) guidelines, these cell lines were considered authenticated. Cells were maintained in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 U/mL penicillin–streptomycin in a humidified atmosphere containing 5% CO₂ at 37 °C. A375-R cells and M14-R were continuously cultured in the presence of 1 µM dabrafenib, which was refreshed every 72 h at each passage. All cell lines were confirmed to be mycoplasma-free (MycoAlert® Mycoplasma Detection Kit, Lonza, Basel, Switzerland) before cryopreservation and used between passages 5 and 15 for all experiments.

TET2 expression was transiently silenced using a TET2-specific small interfering RNA (code SASI_HS02_00328773, Merck Life Science) and Lipofectamine™ 3000 Transfection Reagent (Thermo Fisher Scientific), following the manufacturer’s instructions adapted for a 6-well plate format. Scramble negative control was obtained using MISSION® siRNA Universal Negative Control #1 (code SIC001, Merck Life Science). Cells were seeded 24 h prior to transfection to achieve approximately 60%-70% confluence. For each well of a 6-well plate, 75 pmol of TET2 or SCR siRNA were complexed with 7.5 µL of Lipofectamine™ 3000 in Opti-MEM™ Medium according to the manufacturer’s recommendations for siRNA transfection. The transfection complexes were added to the cells, which were incubated for 24 h under standard culture conditions before further analyses or treatments. The efficiency of TET2 knockdown was confirmed by Western blot analysis.

Cell viability was assessed using the Trypan blue exclusion assay. After treatment, equal volumes of cell suspension and 0.2% (v/v) Trypan blue solution were mixed. Viable cells remained unstained, whereas non-viable cells appeared blue. Cells were counted manually using a Thoma cell counting chamber.

Cells were seeded in 96-well plates and treated with the indicated compounds for the specified times. After treatment, the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) reagent was added to each well and incubated for 4 h at 37 °C to allow formation of formazan crystals. The medium was then removed, and the crystals were dissolved in DMSO. Absorbance was measured at 570 nm using a microplate reader (TECAN Infinite 200 pro), and cell viability was expressed as a percentage relative to untreated controls. To evaluate the influence of glucose availability on ASC-induced cytotoxicity, cells were cultured in RPMI-1640 medium supplemented with glucose at increasing concentrations (2.5, 5, 10, and 20 mM) and subsequently treated with ASC. To assess the involvement of oxidative stress, cells were pretreated with NAC (5 mM) for 4 h prior to ASC exposure, whereas Trolox (50 µM) was administered simultaneously with ASC as a co-treatment. For 2D culture experiments, cells were treated with 0.5 or 1 mM ASC or p-ASC for 72 h. For three-dimensional (3D) spheroid cultures, cells were treated with 0.5, 1, or 2 mM ASC for 5 days.

3D spheroids were generated using 96-well U-bottom plates. Cells were suspended in RPMI-1640 medium (100 µL) supplemented with methylcellulose at a final concentration of 10 mg/mL, obtained by diluting a 100 mg/mL methylcellulose stock solution 1:10 in culture medium. Cells were seeded at a density of 2,000 cells per well and centrifuged for 10 min at 400 × g to promote spheroid formation. Spheroid development was monitored after 24 h using an inverted light microscope (Axio Observer microscope, Zeiss). Once spheroids were formed, treatments were applied (0.5, 1, or 2 mM ASC for 5 days). Cell viability was evaluated using the MTT assay according to the manufacturer’s instructions.

Cells were seeded at a density of 8 × 10⁴ cells/mL in 24-well plates and treated with 0.5 mM ASC or p-ASC for 24 h before the assay. Following treatment, the culture medium was removed and replaced with fresh medium (without FBS) containing 20 µM DCFDA. Cells were incubated for 30 min at 37 °C in the dark. Cells were then washed three times with PBS, and fluorescence was observed with 485 nm excitation and 535 nm emission using a fluorescence microscope (Zeiss Axio Observer).

Mitochondrial membrane potential was assessed using the JC-1 probe, following the same procedure used for reactive oxygen species (ROS) detection. Cells were seeded at a density of 8 × 10⁴ cells/mL in 24-well plates. After 24 h, cells were treated with 0.5 mM ASC or p-ASC for 24 h. Following treatment, the culture medium was removed and replaced with fresh medium (without FBS and Phenol Red) containing 2.5 μM JC-1. Cells were incubated for 30 min at 37 °C in the dark and then washed three times with PBS. Green (monomeric) and red (J-aggregate) fluorescence were acquired separately using a fluorescence microscope (Zeiss Axio Observer), and the red-to-green fluorescence ratio was used as an index of mitochondrial membrane potential.

Intracellular glutathione levels were measured using the GSH/GSSG-Glo™ Assay (Promega, Madison, WI, USA) according to the manufacturer’s instructions, with minor modifications for adherent cells. After treatment with 0.5 mM ASC for 24 h, the culture medium was removed, and cells were lysed by adding 50 µL per well of either Total Glutathione Lysis Reagent or Oxidized Glutathione Lysis Reagent. Plates were shaken for 5 min at room temperature to ensure complete lysis. Subsequently, 50 µL of Luciferin Generation Reagent was added, and plates were incubated for 30 min at room temperature. Then, 100 µL of Luciferin Detection Reagent was added, followed by a 15 min incubation, and luminescence was measured using a microplate luminometer. Background luminescence from wells without cells was subtracted from all measurements to obtain net relative luminescence units (RLU). The GSH/GSSG ratio was calculated using the manufacturer’s instructions.

Cell pellets were resuspended in lysis buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton™ X-100, 0.5% sodium deoxycholate, 0.1% SDS, 10 mM sodium fluoride, 5 mM sodium pyrophosphate, and 2 mM sodium orthovanadate) supplemented with a protease inhibitor cocktail (AMRESCO, M221). Homogenates were sonicated, and protein concentrations were determined using the Lowry method. Protein lysates were mixed with sample buffer (0.125 M Tris-HCl pH 6.8, 4% SDS, 20% glycerol, 10% β-mercaptoethanol, and 0.004% bromophenol blue), separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), and transferred onto nitrocellulose membranes (Bio-Rad). Membranes were incubated with the following primary antibodies (1:1,000 dilution): TET1, TET2, TET3, GLUT1, CAT, SOD2, α-actinin, and β-actin from Cell Signaling Technology. For normalization of Western blotting analysis, α-actinin and β-actin were used as loading controls for proteins resolved on 6.5% and 10%-12% polyacrylamide gels, respectively. To ensure accurate data interpretation, the corresponding loading control is shown alongside each target protein, including when shared within the same panel. Secondary antibodies used were goat anti-rabbit IgG (H + L)–HRP conjugate and goat anti-mouse IgG (H + L)–HRP conjugate from Bio-Rad. Signals were detected using Clarity Max Western ECL Substrate (Bio-Rad) and acquired using a FluorChem imaging system (Alpha Innotech). Densitometric analysis was performed using ImageJ software.

Cells were seeded at a density of 8 × 10⁴ cells/ml on glass coverslips or glass-bottom dishes. At the end of the treatment with 0.5 mM ASC or p-ASC for 24 h, the culture medium was removed, and cells were washed with PBS before fixation. Cells were fixed with formalin solution for 15 min at room temperature and washed three times with PBS. Cells were permeabilized with 0.1% Triton X-100 for 10 min and incubated with RNase for 2 h at 37 °C. For detection of 5-hmC, DNA was denatured by incubating cells with 4 N HCl for 15 min at room temperature, followed by neutralization with 100 mM Tris-HCl (pH 8.5) for 10 min. Cells were then washed with PBS and blocked with 3% goat serum in PBST for 1 h at room temperature. Samples were incubated overnight at 4 °C with rabbit anti-5-hydroxymethylcytosine (5hmC; Active Motif) primary antibody diluted in blocking buffer. After washing with PBST, cells were incubated for 1 h at room temperature with the appropriate Alexa Fluor–conjugated secondary antibodies. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) for 10 min. Coverslips were washed with PBS and mounted on glass slides. Fluorescence images were acquired using a confocal fluorescence microscope (Olympus) and quantified with ImageJ software.

Three independent RNA-seq analyses from publicly available datasets were assessed. First, RNA-seq data from vemurafenib-sensitive and -resistant melanoma cell lines were compared and are available in the GSE203545 dataset. Resistant models were generated through chronic drug exposure as described in the original study^[17]. Samples were grouped according to cell line status (parental vs. vemurafenib-resistant), and only cell line pairs with detectable baseline TET expression in the parental (sensitive) cells were included, while those with negligible expression were excluded to ensure meaningful comparative analyses. Gene expression levels were quantified as fragments per kilobase of transcript per million mapped reads (FPKM), normalizing read counts for both gene length and sequencing. Statistical analysis was performed by a paired t-test because each vemurafenib-resistant melanoma cell line was generated from its corresponding parental cell line. Thus, parental and resistant samples constitute matched pairs rather than independent groups, making a paired analysis the most appropriate approach to assess resistance-associated changes in TET gene expression while accounting for baseline differences among cell lines. Second, gene expression modulation was analyzed in the human metastatic melanoma cell line WM239A overexpressing BRAFV600E using the publicly available dataset GSE117123. In this dataset, cells were treated with the BRAFi encorafenib (LGX818) over a time course ranging from one to three weeks. Prior to analysis, genes with fewer than 60 total reads were excluded. The remaining genes were analyzed using the DESeq2 package, which was used to normalize read counts for library size and estimate gene-wise dispersion parameters. Normalized data were subsequently used for downstream differential TET gene expression analyses, and statistical significance was determined relative to untreated BRAFV600E controls. Finally, to assess correlations between genes of interest in a large cohort of melanoma patients, Spearman’s correlation analysis was performed using the Skin Cutaneous Melanoma (SKCM) dataset from The Cancer Genome Atlas (TCGA), accessed via the GEPIA2 website^[18]. This cohort includes RNA-seq data from over 400 melanoma tumor samples, processed through a standardized pipeline and normalized as Transcripts Per Million (TPM). All available tumor samples were included in the analysis without additional exclusion criteria.

All experiments were performed in n = 3 independent biological replicates. Data are presented as the mean ± standard deviation (SD). Statistical analyses for multiple comparisons were performed by Graphpad Prism 9 using one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test, while two-tailed Student’s t-tests were applied when comparing two groups. A P-value < 0.05 was considered statistically significant.

To investigate the contribution of TET hydroxylases to BRAFi resistance, we generated A375 and M14 mutated BRAF(V600E) melanoma cell lines with acquired resistance to dabrafenib (hereafter referred to as A375-R and M14-R) to compare the abundance of TET1, TET2, and TET3 proteins with their parental counterparts (A375-P and M14-P). Immunoblot analysis revealed an overall reduction in TET protein levels in both resistant cell lines, with the exception of TET3, which was upregulated in A375-R cells [Figure 1A]. Analysis of publicly available RNA-seq datasets demonstrated that TET2 and TET3 genes were significantly downregulated in response to other BRAFi, as observed in melanoma cell lines resistant to vemurafenib [Figure 1B] or treated with encorafenib (LGX818) [Figure 1C]. Collectively, these findings support a role for TET enzymes in the adaptive response to BRAFi therapy.

ASC is a well-established activator of TET catalytic activity^[19] and, when administered at pharmacological doses, exhibits antineoplastic effects^[10]. Importantly, ASC-driven 5hmC accumulation has been shown to rely predominantly on TET2^[20]. Based on this evidence, we investigated how the altered TET expression profile in dabrafenib-resistant A375 cells influences their cytotoxic response to ASC, focusing on the specific role of TET2. Cell viability assays performed after 72 h of exposure to 0.5 or 1 mM ASC demonstrated that A375-R cells were less sensitive to ASC compared with parental A375-P cells [Figure 2A]. This differential sensitivity was confirmed in a 3D culture system by measurement of spheroid viability, where resistant cells retained higher viability following ASC treatment, whereas parental cells exhibited a marked reduction in viability, reaching a 50% decrease at 2 mM ASC compared with the untreated control (CTRL) group [Figure 2B and C]. We then silenced TET2 [Figure 2D] and found that ASC no longer reduced cell viability [Figure 2E]. Consistently, immunofluorescence analyses revealed a marked increase in 5hmC upon ASC treatment in scrambled siRNA–transfected cells (scramble), whereas only minimal 5hmC staining was detected after TET2 knockdown (siTET2) [Figure 2F and G].

High-dose ASC is also known to promote ROS generation^[21,22]. To clarify whether oxidative stress contributes to ASC-mediated cytotoxicity, we tested equimolar concentrations of p-ASC, an oxidation-resistant and more stable ASC derivative. Unexpectedly, p-ASC failed to induce cytotoxicity in either dabrafenib-sensitive or -resistant A375 cells [Figure 3A]. Despite this, p-ASC elevated 5hmC levels to an extent comparable to ASC [Figure 3B], indicating that TET activation alone does not account for ASC-induced cell death. We therefore hypothesized that the distinct biological effects of ASC and p-ASC stem from their different pro-oxidant properties. DCFDA analysis showed a significant increase in ROS following ASC treatment, whereas p-ASC was considerably less effective [Figure 3C]. Interestingly, the analysis of the glutathione antioxidant system showed a lower ratio of reduced (GSH) to oxidized (GSSG) forms - indicative of greater oxidative stress - in parental cells exposed to ASC compared with untreated samples, whereas no change was detected in A375-R [Figure 3D]. In addition, the analysis of the antioxidant enzymes superoxide dismutase 2 (SOD2) and catalase (CAT) revealed higher basal SOD2 expression in resistant cells, supporting their reduced sensitivity to ASC [Figure 3E]. Notably, by using the antioxidant compounds NAC [Figure 3F] or Trolox [Figure 3G] in combination with ASC, we observed that cell viability was completely recovered.

The mechanism underlying ROS production typically depends on the oxidation of high-dose ASC in the extracellular space and its subsequent reduction once inside the cell, which induces a redox imbalance^[22,23]. Because the uptake of oxidized ASC relies on the GLUT1 transporter^[24], we examined its expression in dabrafenib-sensitive and -resistant cell lines. Interestingly, GLUT1 protein levels were lower in A375-R than in A375-P cells [Figure 4A]. Glucose transporter saturation induced by elevated glucose levels attenuated ASC cytotoxicity [Figure 4B], supporting the notion that ASC relies on GLUT-mediated uptake to enter cells. We next investigated whether GLUT1 protein levels are regulated by TET2 by performing gene-silencing experiments. These analyses revealed that GLUT1 protein levels were reduced following TET2 knockdown in both A375-P and A375-R cells [Figure 4C]. Finally, we demonstrated that TET2 silencing attenuates ASC-induced cytotoxicity by altering ROS generation. DCFDA fluorescence increased only in SCR-transfected cells upon ASC exposure, whereas ROS levels remained comparable to untreated controls in TET2-silenced cells, even after ASC treatment [Figure 4D]. Consistent with these findings, mitochondrial depolarization (low red-to-green JC1 ratio) in ASC-treated SCR cells is not evident following TET2 depletion [Figure 4E], further supporting a TET2-dependent mechanism underlying ASC-induced oxidative cytotoxicity. Finally, analysis of RNA-seq datasets from cutaneous melanoma human samples (SKCM-TCGA) showed that TET2, as well as TET1 and TET3, positively correlates with genes encoding GLUT1 (i.e., SLC2A1) and the H₂O₂ detoxifying enzymes CAT and GSH peroxidase (GPX) 7 and 8 [Figure 4F].

To corroborate the results obtained in the A375 cell line, we repeated the key experiments in M14 cells. We confirmed that ASC is less cytotoxic in M14-R than in M14-P cells [Figure 5A], while neither cell line was affected by p-ASC [Figure 5B]. Although both compounds increased 5hmC levels in the M14 pair [Supplementary Figure 1A], the pro-oxidant effect of ASC was particularly pronounced in the parental line [Figure 5C], as reflected by a marked decrease in the GSH/GSSG ratio [Figure 5D]. M14-R also showed higher levels of SOD2 and CAT [Figure 5E]. TET2 silencing reversed the cytotoxic effects of ASC [Figure 5F], the ASC-induced increase in 5hmC levels [Supplementary Figure 1B], ROS generation [Figure 5G], and mitochondrial depolarization [Supplementary Figure 1C]. Consistent with observations in A375 cells, the parental M14 line expressed higher levels of GLUT1 than its resistant counterpart [Figure 5H]. Moreover, increasing glucose concentrations limited ASC-induced cytotoxicity [Figure 5I] and TET2 knockdown effectively downregulated GLUT1 protein levels [Figure 5J].