The human melanoma cell lines, A2058 and A375, were purchased from the American Type Culture Collection and cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (all from Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C in a 5% CO₂ humidified incubator. Vemurafenib (PLX4032) (MedChemExpress, Inc.) resistant A375 melanoma cells (A375R) were generated by adding increasing concentrations of PLX4032 into A375 cells for more than half a year (17). The final concentration of PLX4032 used to maintain A375R was 2 µM. To verify whether A375R cells were resistant to PLX4032, both parental A375 and resistant A375R cells were treated with various concentrations (0.1, 0.5, 1, 5 and 10 µM) of PLX4032 for 48 h, and the cell viabilities were detected by MTT assay. Furthermore, both parental A375 and resistant A375R cells were treated with different concentrations (0, 2 and 5 µM) of PLX4032 for 24 h, and cell apoptosis was assessed by Annexin V and 7-AAD staining.

Cell viability was determined by MTT assay. Cells were seeded into 96-well plates at a density of 5,000 cells/well and cultured in an incubator overnight at 37°C in 5% CO₂. Then various concentrations of PLX4032 (0, 7.5 and 10 µM) and/or IS (0, 20 and 25 µM in A2058 cells; 0, 10 and 20 µM in A375R cells) were added into corresponding wells, and incubated for 48 h at 37°C. For NAC/GSH treatment experiment, cells were pre-treated with 5 mM NAC/1 mM GSH for 1 h, and then treated with PLX4032 (10 µM) and IS (20 µM) for another 18 h. Next, 20 µl MTT solution (Sigma-Aldrich; Merck KGaA) with a concentration of 5 mg/ml was added to every well and incubated at 37°C for 2–4 h. Subsequently, 100 µl DMSO was added to each well to dissolve the formed formazan crystal. Finally, the optical density (OD) was detected at 492 nm using Infinite F200 Pro microplate reader (Tecan Group, Ltd.). The cell viabilities were calculated via the following formula: Cell viability (%)=(mean OD treatment/mean OD control) × 100.

The cell apoptosis rates were determined by PE Annexin V Apoptosis Detection kit I (BD Biosciences) via flow cytometry according to the manufacturer's instructions. Briefly, cells were seeded in 6-well plates at a density of 2×10⁵ cells/well, and cultured in an incubator overnight at 37°C in 5% CO₂. Then PLX4032 (0 and 10 µM) and/or IS (0 and 20 µM) were added into corresponding wells, and incubated for 24 or 72 h at 37°C. Next, the cells were harvested and washed three times with PBS (Gibco; Thermo Fisher Scientific, Inc.), and resuspended with 100 µl 1X binding buffer, and then 5 µl PE Annexin V and 20 µl 7-Amino-Actinomycin (7-AAD) were added into each tube and incubated at room temperature in the dark for 20 min. Finally, the labeled cells were washed with 1X binding buffer, and the data were obtained by a flow cytometer (Attune NxT; Thermo Fisher Scientific, Inc.) and analyzed by FlowJo software v10.0 (Tree Star, Inc.).

A total of 20×10⁴ cells were seeded into each well of a 6-well plate and incubated overnight. Then the cells were treated with vehicle control, IS, PLX4032, and IS+PLX4032 for 24 h. The treated cells were loaded with DCFH-DA probes for 20 min at 37°C, followed by a three-time wash with DMEM without FBS. Finally, the signals were detected by a flow cytometer and analyzed by FlowJo software v10.0.

The treated cells were lysed in RIPA lysis buffer (Beyotime Institute of Biotechnology) containing 10% PMSF (Beyotime Institute of Biotechnology) at 4°C for 15 min with intermittent vortex, and then centrifuged at 4°C at 12,000 × g for 10 min. The supernatants were immediately collected. The protein concentrations were measured by a BCA Protein Assay kit (Life Technology; Thermo Fisher Scientific, inc.), according to the manufacturer's instructions. In addition, 5X loading buffer (Beyotime Institute of Biotechnology) was added into the collected supernatants at a ratio of 1:4, and the mixture was heated at 100°C for 10 min. Total amounts of 30 µg cellular proteins were electrophoresed on 10% SDS-PAGE (Beyotime Institute of Biotechnology), and subsequently transferred onto polyvinylidene difluoride (PVDF) membranes (EMD Millipore), which was followed by blocking with 5% nonfat milk at room temperature for 1 h. Then the membranes were incubated with the primary antibody at 4°C overnight, and followed by an incubating step with a secondary goat anti-rabbit antibody conjugated with horseradish peroxidase (HRP; product code ab6721; dilution 1:4,000; Abcam) for 1 h at room temperature. Finally, the bands were measured via ECL technology (Fujifilm LAS-3000; Fujifilm Corporation). The band intensities were quantified by ImageJ software (v1.37; National Institutes of Health). β-Actin was used as a loading control, and the relative band intensities were normalized to the medium control (ME) group. Primary antibodies against p53 (product code ab32389; dilution 1:1,000), phosphorylated (p)-p53 (product code ab1431; dilution 1:1,000), LC3B (product code ab192890; dilution 1:1,000), Beclin1 (product code ab207612; dilution 1:1,000), Bax (product code ab32503; dilution 1:1,000), Bcl-2 (product code ab32124; dilution 1:1,000), MITF (product code ab232415; dilution 1:1,000), c-Met (product code ab51067; dilution 1:1,000), and β-actin (product code ab8227; dilution 1:1,000) were supplied by Abcam.

The treated cells were stained with the JC-1 fluorescent probe (Beyotime Institute of Biotechnology) following the manufacturer's instructions. The mitochondrial membrane potential was detected by a flow cytometer and analyzed by FlowJo software v10.0. We referred to the study of Chen et al to draw the cut-off line (18).

A2058 or A375R cells were seeded into a 6-well plate at a concentration of 2×10⁵ cells/well, and treated with vehicle, PLX4032, IS, or PLX4032+IS for 24 h. The treated cells were incubated with MitoROS™ 580 (AAT Bioquest, Inc.) working solution for 30 min and washed 3 times with Hanks' solution with 20 mM HEPES buffer (AAT Bioquest, Inc.). The mean fluorescence intensity was detected by a cytometer and analyzed by FlowJo software v10.0.

A total of 1,000 A2058 or A375R cells were seeded into each well of a 6-well plate and treated with vehicle, PLX4032, IS, or PLX40332+IS for 9 days. The medium was changed every 3 days. Finally, the treated cells were fixed with 4% paraformaldehyde for 1 h at 4°C and stained with 0.05% crystal violet solution for 15 min at room temperature.

All data were analyzed by GraphPad Prism 6 (GraphPad Software, Inc.) and presented as the mean ± standard deviation of three independent experiments. For comparison of two independent groups, Student's t-test was used. For comparison of more than two independent groups, one-way ANOVA (one factor) or two-way ANOVA (two factors) and subsequent Tukey's post hoc analysis were applied. P<0.05 was considered to indicate a statistically significant difference.

To verify the effects of IS on BRAF inhibitor-resistant melanoma, the effects of PLX4032 and/or IS on the cell proliferation of two BRAF inhibitor-resistant melanoma cell lines, A2058 (intrinsic resistance) (19) and A375R (acquired resistance, Fig. S1A and B) were first evaluated. In the clonogenic assay, it was revealed that either PLX4032 or IS alone could inhibit the proliferation of both A2058 and A375R cells. Moreover, the combination of PLX4032 and IS revealed increased inhibition of cell proliferation compared to PLX4032 or IS alone (Fig. 1B and C). It was also demonstrated that in the MTT assay, various concentrations of PLX4032 or IS alone could decrease the cell viability of both A2058 and A375R cells, and combined IS and PLX4032 further reduced the cell viability of both A2058 and A375R cells (P<0.05, Fig. 1D and E). Moreover, IS and PLX4032 exhibited synergetic effects in inhibiting the proliferation of both A2058 and A375R cells (Fig. 1F and G).

The effects of PLX4032 and/or IS on the apoptosis of BRAF inhibitor-resistant cells were further examined by staining with Annexin V and 7-AAD. As revealed in Fig. 2A and B, cells treated with either PLX4032 or IS alone exhibited a higher apoptosis rate than ME-treated cells (P<0.05). In addition, the highest apoptosis rate was observed in cells treated with both PLX4032 and IS (P<0.05, Fig. 2A and B). In addition, the highest expression of autophagy-related proteins, LC3B and Beclin1, was revealed in the PLX4032+IS group in both A2058 and A375R cells (Fig. 2C and D).

Having revealed the effect of IS on cell proliferation, apoptosis and autophagy in BRAF inhibitor-resistant melanoma cells, the underlying mechanisms were next determined. As indicated in Fig. 3A and B, IS significantly decreased the percentages of JC-1 aggregates and increased the proportion of JC-1 monomers in A2058 cells, indicating that IS could induce mitochondrial depolarization (P<0.05). However, PLX4032 had little effect on mitochondrial membrane potential (P>0.05, Fig. 3A and B). In addition, it was revealed that the combination of PLX4032 and IS promoted the depolarization of mitochondria to a greater extent than IS or PLX4032 alone (P<0.05, Fig. 3A and B).

The expression of two apoptotic regulators in the Bcl-2 family, Bax and Bcl-2, which are co-localized on the outer membrane of mitochondria and regulate the opening of mitochondrial voltage-dependent anion channels, thus moderating the mitochondrial membrane potential and release of cytochrome c, was then examined (20). As revealed in Fig. 3C, IS with or without PLX4032 significantly downregulated Bcl-2 expression and upregulated Bax levels in A2058 cells.

Given that p53 participates in apoptosis by directly interacting with and neutralizing Bcl-2 and Bcl-xL in mitochondria, and directly activating Bax in the cytosol (21), it was hypothesized that IS led to apoptosis by p53-dependent inhibition of Bcl-2, activation of Bax and subsequent mitochondrial membrane permeabilization. As anticipated, PLX4032 did not activate p53, while IS or IS+PLX4032 significantly upregulated the expression of p-p53 in A2058 cells (Fig. 3D).

Since higher ROS generation has been reported to sensitize resistant cells to BRAF inhibitors, the ROS levels in all four groups, i.e. ME, PLX4032, IS, PLX4032+IS, were then assessed. As anticipated, higher ROS levels were detected in the IS and PLX4032+IS groups than in the ME or PLX4032 alone group (P<0.05, Fig. 4A). Given the role of IS on the mitochondrial membrane, it was hypothesized that IS induced-ROS was generated from mitochondria. To verify this hypothesis, mitochondrial ROS levels were assessed in all groups in both A2058 and A375R cells. In accordance with total ROS generation, IS with or without PLX4032 significantly increased mitochondrial ROS levels (P<0.05; Fig. 4B).

To further confirm the role of ROS in the IS-improved response to PLX4032, NAC and GSH, ROS scavengers, were then used to eliminate ROS secretion, and it was detected whether combination of PLX4032 and IS inhibition of cell viability could be reversed. As indicated in Fig. 4C and D, the cell viability of NAC/GSH+PLX4032+IS-treated cells was significantly higher than that of PLX4032+IS-treated cells (P<0.05).

MITF and c-Met are well-known factors that contribute to BRAF inhibitor resistance (22–24). Furthermore, c-Met is a direct transcriptional target of MITF in melanocytes and melanoma cells (25). Therefore, the effects of IS on the expression of both MITF and c-Met, were then verified. As demonstrated in Fig. 5A-D, decreased mRNA and protein levels of MITF and c-Met were observed in the IS and IS+PLX4032 groups than in the ME group in both A2058 and A375R cells (P<0.05). However, PLX4032 alone could not markedly downregulate MITF expression (Fig. 5A-D). PLX4032 significantly increased c-Met expression in A2058 cells compared to the ME group, while it significantly decreased c-Met expression in the A375R cells (Fig. 5A-D). To verify that IS induced downregulation of MITF and c-Met by increasing ROS production, H₂O₂, exogenous ROS, was used to promote ROS generation and it was revealed that H₂O₂ could significantly downregulate MITF and c-Met expression in both A2058 and A375R cells (Fig. 5E). These results indicated that IS may improve the response to PLX4032 partially by increasing ROS production and subsequently inhibiting MITF and c-Met expression in resistant melanoma cells.