Introduction

Melanoma, a type of skin cancer originates in the pigment-producing melanocytes of the skin. Although it accounts for about 1% of skin cancers, it causes the majority of skin cancer-related deaths [https://www.cancer.org/cancer/melanoma-skin-cancer/about/key-statistics.html]. The 5-year survival rate for patients with metastatic melanoma is around 30% [1]. 50% of melanomas harbor activating mutations in v-raf murine sarcoma viral oncogene homolog B1 (BRAF) oncoprotein with over 90% of these mutations being V600E (valine at position 600 substituted by glutamic acid) [2]. The combination of dabrafenib, a BRAF inhibitor and trametinib, a MEK inhibitor is Food and Drug Administration (FDA)-approved targeted therapy for patients with BRAF-mutant melanoma [3]. Chronic treatment with these inhibitors leads to the development of acquired resistance as the clinical response to these therapies drops [4]. Constitutive activation of mitogen-activated protein kinase pathway (MAPK) and phosphoinositide 3-kinase (PI3K)/Akt pathway has been reported to contribute to resistance in melanoma [5–9]. Previous studies from our group have shown that BRAF-mutant melanoma cells resistant to dabrafenib and trametinib maintained p-Akt levels compared to their drug-sensitive counterparts [10].

Reactive oxygen species (ROS) are signaling molecules for normal biologic processes. An excess of ROS damages cellular organelles and DNA [11]. ROS are generated by partial reduction of oxygen comprised of radical and non-radical oxygen species such as superoxide anion (O2−), hydrogen peroxide (H2O2), and hydroxyl radical (HO•). ROS is mainly generated endogenously in cells during oxidative phosphorylation in the mitochondria or in response to exogenous stimuli such as xenobiotics, ionizing radiations, and nutrients [12,13]. An imbalance in ROS and antioxidants, which serve as a defense mechanism to ROS, has been implicated in cancer [11,14,15]. ROS has been implicated in the occurrence and development of skin carcinogenesis, most importantly melanoma [16,17]. BRAF inhibition induces and renders melanomas addictive to oxidative phosphorylation via peroxisome proliferator-activated receptor gamma coactivator 1-alpha and melanocyte-inducing transcription factor [18]. Previous studies from our group and other have demonstrated that elevated levels of ROS mediate resistance to BRAF and MEK inhibitors in melanoma. ROS levels are augmented on acute and chronic treatment with BRAF and MEK inhibitors [19–22]. The PI3K/Akt pathway is also subjected to redox regulation in presence of excess ROS [23].

The PI3K proteins are heterodimeric lipid kinases. These PI3Ks are divided into three classes based on their structure and specificity. The class I PI3Ks are further subdivided into two subclasses- PI3Kα, PI3Kβ, and PI3Kδ into class IA and PI3Kγ into class IB. Class IA PI3Ks are composed of one p110 catalytic subunit encoded by PIK3CA (p110α), PIK3CB (p110β), or PIK3CD (p110δ) and a regulatory p85 subunit. PI3Kγ consists of a p110γ subunit encoded by PIK3CG and one p101 regulatory subunit [24]. Upon ligand binding to cognate RTKs, PI3K is recruited to the cell surface and catalyzes the conversion of secondary messenger molecule phosphatidylinositol 4,5-bisphosphate (PIP2) to phosphatidylinositol 3,4,5-triphosphate (PIP3). This conversion activates several downstream effectors such as PDK and AKT serine/threonine kinases which in turn activate several downstream signaling pathways. This pathway plays a major role in tumorigenesis in malignant melanoma [25].

PI3Ks are promising anticancer targets and many small molecule inhibitors have been developed as cancer therapy. The two classes of PI3K inhibitors are pan inhibitors (targeting the four isoforms of class I PI3Ks) and isoform-specific inhibitors [25]. Even though several PI3K inhibitors are being investigated in clinical trials, only a few have been FDA approved due to drug-related toxicities [26]. PI-103 is a pan PI3K inhibitor with strong activity against PI3Kα isoform. PI-103 did not enter clinical trials due to limited solubility and extensive metabolism [27]. A boron-containing bioisostere of PI-103 offered improved bioavailability [28]. ROS-induced drug release (RIDR)-PI-103 is a prodrug of PI-103, which in presence of high ROS releases the biologically active component, PI-103 [29,30].

In this study, we have tested the efficacy of RIDR-PI-103 on trametinib and dabrafenib-resistant (TDR) melanoma cells. We observed that RIDR-PI-103 inhibited cell proliferation in TDR cells as compared to control melanocytes. RIDR-PI-103 impaired PI3K/Akt signaling in a dose-dependent manner in TDR cells along with inhibition of downstream p-ribosomal protein 6 (S6). We demonstrate that RIDR-PI-103 in combination with glutathione, ROS scavenger rescues cell proliferation of TDR cells. RIDR-PI-103 in combination with t-butyl hydrogen peroxide (TBHP), ROS inducer inhibits cell proliferation. We demonstrate that RIDR-PI-103 in a dose-dependent manner inhibited the migration of TDR cells. These findings provide mechanistic insight and rationale for targeting drug-resistant melanomas with RIDR-PI-103.

Materials and methods Reagents

RIDR-PI-103 was synthesized as described [29]. PI-103 (Cat# S1038) was obtained from Selleckchem (Houston, Texas, USA). All drugs were dissolved in dimethyl sulfoxide (DMSO) (Fisher Scientific, Waltham, Massachusetts, USA) at desired concentrations and stored at −20°C. Glutathione was obtained from Fisher Scientific (Cat# BP25211). N-acetyl cysteine (NAC) was obtained from Fisher Scientific (Cat# AAA1540936). TBHP was obtained from Fisher Scientific (Cat# AC180340050). Dabrafenib (Cat# D-5699) and trametinib (Cat# T-8123) were purchased from LC Laboratories (Woburn, Massachusetts, USA).

Cell lines and culture conditions

TDR cells were generated as previously described [10]. A375 cells were cultured in Dulbecco’s Modified Eagle Medium (cat# MT15017CV; Corning, New York, USA); WM115 cells were cultured in MEM (cat# MT10009CV; Corning); WM983B cells were cultured in RPMI-1640 medium (cat# MT10040CM; Corning) supplemented with 10% (vol/vol) fetal bovine serum (FBS) (cat# 89510-188; VWR) and 1% penicillin–streptomycin solution 100× (cat# 30-002-CI; Corning). Cultured cells were incubated in a humidified incubator at 37°C with 5% (vol/vol) CO2 and 95% (vol/vol) air. A375 TDR cell lines were maintained in 250 nM dabrafenib and 12.5 nM trametinib. WM983B TDR cells were maintained in 2.4 µM dabrafenib and 500 nM trametinib. WM115 TDR cells were maintained in 800 nM dabrafenib and 200 nM trametinib. 1788C and 1789B melanocytes were kindly provided by Dr. Zalfa Abdel-Malek from the University of Cincinnati, College of Medicine.

Transient transfection

5 × 104 cells were plated and transfected with control empty vector plasmid (pECE; Addgene: 26453) and constitutively active AKT plasmid (myrAkt delta4-129; Addgene: 10841). In separate experiments, the above cells were transfected with control empty vector plasmid (LZRS-Rfa; Addgene: 31601) and constitutively active MEK plasmid (L1E-1; Addgene: 21193) using lipofectamine 2000 for 48 h. Cell lysates were collected and analyzed for AKT and MEK pathway activation using western blotting.

Cell proliferation and long-term growth assays MTT assay

Growth kinetics of WM115 TDR, WM983B TDR, A375 TDR, 1789C melanocytes, and 1788B melanocytes was determined by (3-(4,5-dimethy;thiazol-2-yl)-2,5-diphenyltetrazolium bromide) (MTT) assay. 1 × 104 cells/wells were seeded in 96-well plates in quadruplicate. After 72 h of treatment with vehicle, RIDR-PI-103 (2.5 µM, 5 µM, and 10 µM) or PI-103 (5 µM and 10 µM), the media was substituted with 0.5 mg/mL MTT solution (Sigma Aldrich, St. Louis, Missouri, USA) for 4 h and absorbance was read at 570 nm using SPECTRAmax PLUS Microplate Spectrophotometer Plate Reader (Molecular Devices Incorporation, San Jose, California, USA) and expressed as the mean of quadruplicate relative to vehicle (DMSO) control together with SEM. For experiments with glutathione, briefly 1 × 104 cells/wells were seeded in 96-well plates in quadruplicate. Cells were treated with glutathione (100 µM) alone or in combination with RIDR-PI-103 (2.5 µM, 5 µM, and 10 µM) for 72 h, and proliferation was assessed as mentioned above. For experiments with NAC, cells were plated and treated with NAC (50 mm) alone or in combination with RIDR-PI-103 (2.5 µM, 5 µM, and 10 µM) for 72 h and proliferation was assessed as mentioned above. For experiments with TBHP, cells were plated and treated with TBHP (WM115 TDR and WM983B TDR – 20 mM and A375 TDR – 50 mM) alone or in combination with RIDR-PI-103 (2.5 µM) for 72 h and proliferation was assessed as mentioned above. In other experiments, equal number of WM115 TDR cells were plated and transfected with control empty vector plasmid (pECE), constitutively active AKT plasmid (myrAkt delta4-129), control empty vector plasmid (LZRS-Rfa), constitutively active MEK plasmid (L1E-1) for 24 h and treated with 5 µM RIDR-PI103 for another 24 h and MTT assay was performed as described above. The experiment was repeated twice and data were plotted as mean values with respect to that of control plasmids.

BrdU cell proliferation assay

Cell proliferation of WM115 TDR, WM983B TDR, A375 TDR cells was analyzed using BrdU cell proliferation assay kit from Cell Signaling Technology (cat# 6813; Danvers, Massachusetts, USA). Briefly, equal number of TDR cells (2 × 104 cells/well) were plated and treated with 0–60 µM RIDR-PI 103 for 48 h. After 48 h, media was removed and 10X BrdU solution was added to the wells and incubated for 1 h in the incubator. Then, 100 µL/well Fixing solution was added and cells were incubated at room temperature for 30 min. Further, 1X detection antibody solution and 1X horse radish peroxidase (HRP)-conjugated secondary antibody solution were added as per manufacturer’s protocol. In the final step, 100 µL TBP solution was added followed by stop solution, and the plates were read at 450 nm using SPECTRAmax PLUS Microplate Spectrophotometer Plate Reader (Molecular Devices Incorporation) and the values were expressed as the mean of triplicate relative to vehicle (DMSO) control together with SEM.

Crystal violet assay

TDR cells were seeded density of 2 × 104 cells/well in 12-well plates (Falcon) in triplicates. Complete media containing vehicle or RIDR-PI-103 (5 µM and 10 µM) was replenished every 2 days. On day 15, cells were stained with 0.5% crystal violet. The cultures were exposed to crystal violet solution for 2 min and washed with water to remove excess stains. The intensities were measured using an Odyssey infrared system (LI-COR, Lincoln, Nebraska, USA). The experiments were repeated at least twice. For low-dose RIDR-PI-103 experiments, 1.5 × 104 cells/well were seeded in a 24-well plate in quadruplicate. Complete media containing vehicle or RIDR-PI-103 (1 µM, 2.5 µM, and 5 µM) was replenished every 2 days. On day 15, cells were stained as mentioned above.

Matrigel growth assay

Three-dimensional (3D) growth assays were performed in growth factor-reduced Matrigel (cat# CB-40230; BD Biosciences, San Jose, California, USA). Ninety-six-well plates were coated with 70 µL of Matrigel/well in duplicate. TDR cells (2 × 104/well) were plated and incubated at 37°C for 24 h. Cells were treated with vehicle (DMSO) or RIDR-PI-103 (2.5 µM, 5 µM, and 10 µM) every alternate day. After 21 days of incubation, cells were visualized, and photographs were captured from five random fields under phase contrast microscope (10x magnification, Nikon, Melville, New York, USA). The areas of the cells were measured by ImageJ and represented as mean areas normalized to DMSO control.

Western blot analysis

Cells were lysed in RIPA buffer (Cat# BP-115; Thermo Fisher Scientific) supplemented with protease and phosphatase inhibitors (Cat# 88669; Thermo Fisher Scientific). Protein concentration was determined using the Pierce BCA Protein Assay Kit (Cat# 23225; Thermo Scientific) according to manufacturer’s instructions. A minimum of 15 µg of protein was used for each sample. Immunoblotting was performed using the following primary antibodies from Cell Signaling Technology: p-AktS473 (Cat# 4060), Akt (Cat# 9272), p-S6(240/244) (Cat# 2215), p-S6(235/236) (Cat# 4858), S6 (Cat# 2217), and β-Actin (Cat# 4970). The primary antibody (1:1000 dilution) was added to the nitrocellulose and incubated overnight at 4°C with gentle agitation. Blots were then washed with 1x Tris Buffered Saline with Tween (TBST) (three washes, 10 min each) and incubated with anti-rabbit IgG-HRP (cat# SC-2357; 1:2000 dilution; Santa Cruz Biotechnology) for 1 h at room temperature. Blots were then washed with 1X TBST (three washes, 10 min each). Protein signals were detected using Pierce ECL Western blotting substrate (Cat# 32106; Thermo Fisher Scientific). At least two independent repeats were performed.

Cell migration assay

The transwell migration chambers were soaked in complete media for 4 h prior to start of the experiment. Complete media was added to the lower chambers and upper chambers were placed in the wells. 2 × 104 cells/well of WM115 TDR, A375 TDR, and WM983B TDR were counted and added to the upper chambers of the transwells in serum-free media along with vehicle or RIDR-PI-103 (2.5 µM, 5 µM, and 10 µM) and incubated for 24 h. FBS (10%) was used as chemoattractant and added in lower chambers. After 24 h, migrated cells in the lower chamber were stained with 0.5% crystal violet and images were captured from five areas under phase contrast microscope at 10x magnification. The intensities from the captured areas were measured using ImageJ and expressed as % of control and represented graphically.

Trypan blue exclusion assay

Equal number of TDR cells were seeded and treated with vehicle or RIDR-PI-103 (2.5 µM, 5 µM, and 10 µM) for 48 h in 60 mm plates. Cells were collected and centrifuged to get a cell pellet. The cell pellet was resuspended in 1 mL of 1X Phosphate Buffered Saline (Cat# 21-040-CV; Thermo Fisher Scientific). From this, 0.1 mL of the cell suspension was added to 0.1 mL of 0.4% trypan blue solution and incubated for 5 min at room temperature. The cells were immediately counted and cell viability was determined using the formula:

% viable cells=[1.00−(Number of dead cells/Number of total cells)]×100

Statistical analysis

Data are shown as the mean ± SEM and representative of at least three independent experiments unless otherwise indicated. One-way analysis of variance followed by Tukey’s test for post hoc analysis was used for comparisons between more than two groups. Statistical analysis was performed using GraphPad Prism software. P < 0.05 was considered to indicate a statistically significant difference.

Results ROS-induced drug release-PI-103 inhibits growth of BRAF-mutant TDR cells

We evaluated the effects of RIDR-PI-103 on cell proliferation in WM115 TDR, WM983B TDR, A375 TDR, 1789C melanocytes, and 1788B melanocytes at 5 µM and 10 µM concentration using 72-h MTT assay. Additionally, we assessed the effects of PI-103 on 1789C and 1788B melanocytes at 5 µM and 10 µM. Type C melanocytes were obtained from lightly pigmented skin while type B melanocytes were obtained from a dark pigmented skin. We assessed the effects of RIDR-PI-103 only on TDR cell lines since these resistant cells demonstrated significant increase in ROS levels compared to the drug-sensitive parental cells [20]. RIDR-PI-103 at 5 µM did not significantly inhibit proliferation of melanocytes compared to DMSO group. PI-103 significantly inhibited proliferation in 1789C and 1788B melanocytes at 5 µM compared to melanocytes treated with the same concentration of RIDR-PI-103. When melanocytes were treated with 10 µM of RIDR-PI-103 or PI-103, both the parent molecule PI-103 and pro-drug RIDR-PI-103 inhibited the proliferation of melanocytes (Fig. 1).

Fig. 1:

Effects of RIDR-PI-103 and PI-103 on melanocytes. 1789C and 1788B melanocytes were seeded in 96-well plates and treated with PI-103 or RIDR-PI-103 (RIDR) at 5 µM/10 µM for 72 h. Cells were treated with MTT and absorbance was read at 570 nm. Error bars: SEM. **P < 0.01, ***P < 0.001 vs. DMSO; ###P < 0.001, ####P < 0.0001 vs. ANOVA, analysis of variance; RIDR-PI-103 treatment group (one-way ANOVA followed by Tukey’s test for post hoc analysis). RIDR, ROS-induced drug release; ROS, reactive oxygen species.

TDR cells treated with RIDR-PI-103 had significantly inhibited proliferation at 5 µM and further inhibition was observed at 10 µM (Fig. 2a). These data indicate that RIDR-PI-103 has increased specificity in TDR cells relative to melanocytes. RIDR-PI-103 significantly reduced cell growth in TDR cells in a dose-dependent manner compared to DMSO (Fig. 2b and S1, Supplemental Digital Content 1, https://links.lww.com/ACD/A491) as measured by crystal violet staining. RIDR-PI-103 was effective at concentration as low as 2.5 µM in WM115 TDR and WM983B TDR cells (Fig. S2, Supplemental Digital Content 2, https://links.lww.com/ACD/A492). We next examined the growth potential of TDR cells in presence of RIDR-PI-103 at 2.5 µM, 5 µM, and 10 µM on a basement membrane of matrigel. WM983B TDR cells treated with DMSO exhibited aggressive growth. Treatment with RIDR-PI-103 inhibited growth on matrigel basement membrane in a dose-dependent manner (Fig. 2c). We also observed a decrease in cell viability for WM115 TDR and A375 TDR cells at 48 h using increasing concentration of RIDR-PI-103 using cell viability using the trypan blue exclusion assay (Fig. S3, Supplemental Digital Content 3, https://links.lww.com/ACD/A493). In addition, we analyzed the effect of RIDR-PI-103 (0–60 µM) on BRAF and MEK inhibitor-resistant melanoma cell lines cell proliferation using BrdU assay. Our data indicated that RIDR-PI-103 inhibited WM115 TDR, WM983B TDR, and A375 TDR cell proliferation in a dose-dependent manner (Fig. S4, Supplemental Digital Content 4, https://links.lww.com/ACD/A494). We observed a more pronounced effect of RIDR-PI-103 in the MTT, crystal violet, and Matrigel growth assays compared to the BrdU cell proliferation which could indicate RIDR-PI-103 primary effect is on cell viability and not cell division.

Fig. 2:

RIDR-PI-103 inhibits growth of TDR cells. (a) WM983B TDR, WM115 TDR, and A375 TDR cells were seeded in 96-well plates and treated with increasing concentrations of RIDR-PI-103 (RIDR) for 72 h. Cells were treated with MTT and absorbance was read at 570 nm. Data reflect three independent repeats and are presented as mean ± SEM. Error bars: SEM. ****P < 0.0001 vs. DMSO (one-way ANOVA followed by Tukey’s test for post hoc analysis). (b) TDR cells were plated and treated with media containing DMSO or RIDR (5 µM or 10 µM). Media containing drugs was replenished every second day and stained with 0.5% crystal violet on Day 15. Quantitation of intensities for the wells is represented as mean. Error bars: SEM. ***P < 0.001, ****P < 0.0001 vs. DMSO (one-way ANOVA followed by Tukey’s test for post hoc analysis). (C) TDR cells were seeded on a Matrigel basement membrane and treated with indicated treatment every second day. Pictures of wells were captured on day 21 (Top panel) at 10x magnification. Quantification of mean area of acini (Bottom panel). Error bar: SEM. ***P < 0.001, ****P < 0.0001 vs. DMSO (one-way ANOVA followed by Tukey’s test for post hoc analysis). ANOVA, analysis of variance; RIDR, ROS-induced drug release.

ROS-induced drug release-PI-103 inhibits PI3K signaling and downstream S6 signaling

We examined the effect RIDR-PI-103 at 2.5 µM, 5 µM, and 10 µM on signaling pathway in TDR cells by performing a series of western blots. We have previously established that TDR cells maintain p-Akt levels compared to their drug-sensitive counterparts [10]. In line with those results, we observed the presence of p-Akt in TDR cells treated with DMSO control. In A375 TDR cells, we did not observe a statistical reduction in phosphorylation of Akt (S473) at 2.5 µM RIDR-PI-103, while we see a dose-dependent decrease in p-Akt (S473 levels) with higher concentrations of the prodrug. For WM115 TDR and WM983B TDR cells, we observed a prominent reduction in p-Akt (S473) levels staring at 2.5 µM and p-Akt was inhibited further at 5 µM and 10 µM of RIDR-PI-103. Ribosomal protein S6 is a component of the 40S small ribosomal subunit and plays an important role in protein translation, ribosome biogenesis, cell proliferation and growth, apoptosis, glucose homeostasis, and DNA damage in cancer cells [31–33]. Activation of S6 has been implicated in various malignancies such as melanoma, breast cancer, glioblastoma, head and neck squamous cell carcinoma, and non-small cell lung cancer [34]. It has been observed that melanoma cells resistant to BRAF and MEK inhibitors maintain phosphorylated S6 levels [35]. We observed a dose-dependent inhibition of p-S6 at sites Ser240/244 and Ser235/236 when all three TDR cells were treated with RIDR-PI-103 (Fig. 3). Since RIDR-PI-103 targets the PI3K enzyme and downstream the AKT pathway, we examined if constitutive activation of AKT signaling could overcome the inhibitory effect of RIDR-PI-103 in melanoma cells. In addition, we were also interested to check the effect of MEK constitutive activation on RIDR-PI-103 mediated signaling. Since only WM115 TDR cell line was amenable to transient transfection as indicated by increased p-Akt and p-Erk (Fig S5 A, C, Supplemental Digital Content 5, https://links.lww.com/ACD/A495), we performed an MTT assay where WM115TDR cells transfected with using AKT or MEK constitutively active constructs were treated with 5 µM RIDR-PI-103 as described in materials and method section. The data showed that RIDR-PI-103 significantly inhibited cell growth in WM115TDR cell lines transfected with control plasmids, respectively. This effect was rescued in presence of constitutively active AKT but not MEK constitutively active plasmid (Fig S5, Supplemental Digital Content 5, https://links.lww.com/ACD/A495).

Fig. 3:

RIDR-PI-103 inhibits p-Akt and p-S6 in a dose-dependent manner. (a) Immunoblots of WM983B TDR, WM115 TDR, and A375 TDR cells were treated for 24 h with RIDR-PI-103 (RIDR; 2.5 µM, 5 µM or 10 µM) or vehicle (DMSO). Whole-cell lysates from the cell lines were analyzed for western blot using indicated antibodies. Actin served as the loading control. (b) Band intensities for p-Akt (S473), p-S6 (240/244), and p-S6 (235/236) were quantified using ImageJ software and normalized to intensity for total protein. Two independent repeats were performed. Error bars: SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. WM983B TDR DMSO; #P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001 vs. WM115 TDR DMSO; $P < 0.05, $$P < 0.01, $$$P < 0.001 vs. A375 TDR DMSO (one-way ANOVA followed by Tukey’s test for post hoc analysis). ANOVA, analysis of variance; RIDR, ROS-induced drug release; ROS, reactive oxygen species.

Addition of glutathione to ROS-induced drug release-PI-103 rescues cell proliferation in TDR cells

Glutathione is an antioxidant tripeptide consisting of cysteine, glycine, and glutamic acid. We examined the effects of glutathione alone or in combination with RIDR-PI-103 on TDR cell proliferation using the MTT assay. Glutathione alone had no significant effect on proliferation of all three TDR cell lines. RIDR-PI-103 alone inhibited cell proliferation in TDR cells in a dose-dependent manner. We observed a significant rescue in TDR cell proliferation when glutathione was combined with RIDR-PI-103 at 5 µM compared to RIDR-PI-103 (5 µM) only group. For WM983B TDR cells we observed the rescue in cell proliferation induced by RIDR-PI-103 even at 10 µM RIDR-PI-103 while this effect was not observed in A375 TDR and WM115 TDR cells (Fig. 4). We also assessed the effect of another antioxidant, NAC alone or in combination with RIDR-PI-103 on these TDR cells. NAC alone did not have any effect on the proliferation of A375 TDR while it significantly inhibited proliferation in WM115 TDR and WM983B TDR cells compared to control group. We observed rescue in cell proliferation for A375 TDR cells when NAC was combined with RIDR-PI-103 at 5 µM concentration as compared to A375 TDR cells treated with RIDR-PI-103 alone. While NAC itself inhibited cell proliferation in WM983B TDR cells, we also observed a significant inhibition in cell proliferation between NAC treatment group vs NAC + RIDR-PI-103 (10 µM) group. A significant rescue in WM983B TDR cell proliferation was observed in NAC + RIDR-PI-103 (10 µM) group compared to RIDR-PI-103 group (10 µM). Since NAC strongly inhibited proliferation in WM115 TDR cells, interpretation of data for this cell line is difficult (Fig. S6, Supplemental Digital Content 6, https://links.lww.com/ACD/A496). These data indicated that endogenous ROS levels affect the activation of RIDR-PI-103.

Fig. 4:

RIDR-PI-103 in combination with glutathione rescues TDR cell proliferation. TDR cells (WM983B, WM115, and A375) plated for 24 h and treated with DMSO, RIDR-PI-103 (RIDR; 2.5 µM, 5 µM, or 10 µM) treated alone or in combination with glutathione (100 µM) and treated with MTT. The graphs are represented as mean. Error bars: SEM. *P < 0.05, **P < 0.01, and ****P < 0.0001 vs. the group treated with RIDR-PI-103 only at the corresponding concentration (one-way ANOVA followed by Tukey’s test for post hoc analysis). ANOVA, analysis of variance; RIDR, ROS-induced drug release; ROS, reactive oxygen species.

TBHP in addition to ROS-induced drug release-PI-103 inhibits cell proliferation in TDR cells

TBHP, an organic compound is used as ROS inducer. Since TBHP is more stable than hydrogen peroxide, we used TBHP in our experiments instead of hydrogen peroxide to augment ROS levels. We examined the effects of TBHP alone or in combination with RIDR-PI-103 on TDR cells using MTT cell proliferation assay. TBHP alone was not toxic to TDR cells and did not significantly inhibit proliferation in all three TDR cell lines. We observed a significant inhibition in WM115 TDR and A375 TDR cell proliferation when TBHP was added with RIDR-PI-103 at 2.5 µM compared to cells treated only with RIDR-PI-103. For WM983B TDR cells, we did not observe any inhibition in cell proliferation when TBHP was added with 2.5 µM RIDR-PI-103 (Fig. 5). This data demonstrates that augmentation of ROS levels in TDR cells inhibits cell proliferation in two out of the three TDR cell lines.

Fig. 5:

RIDR-PI-103 in combination with TBHP inhibits TDR cell proliferation. TDR cells (WM983B, WM115, and A375) were plated for 24 h and treated with DMSO, RIDR-PI-103 (RIDR; 2.5 µM) treated alone or in combination with TBHP (WM115 and WM983B TDR – 20 mm; A375 TDR – 50 mm) and treated with MTT. The graphs are represented as mean. Error bars: SEM. *P < 0.05 vs. RIDR 2.5 µM (one-way ANOVA followed by Tukey’s test for post hoc analysis). ANOVA, analysis of variance; RIDR, ROS-induced drug release; ROS, reactive oxygen species.

ROS-induced drug release-PI-103 suppresses TDR cell migratory potential

We assessed the cell migration of TDR cells when treated with increasing concentrations of RIDR-PI-103. The TDR cells and RIDR-PI-103 (2.5 µM, 5 µM, and 10 µM or control) were added to the upper chambers of the transwell under serum starvation and allowed to migrate for 24 h to the lower chambers containing 10% FBS. Migration of TDR cells treated with 2.5 µM, 5 µM, and 10 µM of RIDR-PI-103 was significantly reduced at 24 h (Fig. 6). We also performed cell migration for 48 h and no differences were observed between 24 h and 48 h migration when treated with RIDR-PI-103 (data not shown).

Fig. 6:

RIDR-PI-103 inhibits migration in TDR cells. TDR cells (WM983B, WM115, and A375) treated with DMSO, RIDR-PI-103 (RIDR; 2.5 µM, 5 µM or 10 µM) were seeded in upper chamber of transwell in serum-free media for 24 h. 10% FBS was added as chemoattractant in the lower chamber. After 24 h, the migrated cells were stained, and images were captured (Top panel) at 10x magnification. The images were quantified using Image J and presented as bar graphs (Bottom panel). Error bars: SEM. **P < 0.01, ****P < 0.0001 vs. DMSO (one-way ANOVA followed by Tukey’s test post hoc analysis). ANOVA, analysis of variance; FBS, fetal bovine serum; RIDR, ROS-induced drug release; ROS, reactive oxygen species.

Discussion

The PI3K/Akt pathway serves as a therapeutic target in drug-resistant melanoma [36]. The two most common events that attribute to PI3K/Akt pathway activation in melanoma are mutations in the N-Ras oncogene (15–20%) and loss of expression or function in phosphatase and tensin homolog (PTEN) (20–30%). There is a strong association between complete loss of PTEN and BRAF mutations [37–39]. Only five PI3K inhibitors – Copanlisib (Aliqopa), Idelalisib (Zydelig), Alpelisib (PIQRAY), Duvelisib (COPIKTRA), and Umbralisib (Ukoniq) have been FDA approved to date [26]. Despite numerous PI3K inhibitors in preclinical and clinical trials, only a handful have been granted FDA approval as there are several factors that have limited the successful development of these inhibitors.

The PI3K pathway plays a role in basic cellular processes throughout the body including autophagy, protein translation, and metabolism. Owing to the ubiquitous role of this pathway, drugs designed to inhibit PI3K pathway are associated with toxicities [40]. A major challenge in PI3K pathway inhibitor development is that drugs fail to achieve optimal target blockade in tumors. Common toxicities observed with pan-PI3K inhibitors are fatigue, hyperglycemia, and rash. Some toxicities are also isozyme specific. While PI3Kδ inhibitors are associated with myelosuppression, gastrointestinal effect, and transaminitis, the PI3Kα inhibitors are associated with side effects such as rash and hyperglycemia. Due to these toxicities, it is challenging to combine PI3K inhibitors with other small molecule inhibitors [41]. PI3K inhibition blocks the action of insulin, which in turn prevents glucose uptake in adipose tissue and stimulates glycogen breakdown in the liver. This ultimately leads to hyperglycemia, since genes encoding for glycolytic enzymes are transcribed by PI3K/Akt [42]. PI-103 is also associated with these toxicities and hence we have designed a prodrug, RIDR-PI-103 for this pan-PI3K inhibitor, which is activated in the presence of high oxidative stress in the cells. The goal of this prodrug is to circumvent toxicities observed with systemic inhibition of PI3K signaling.

The phenol in the structure of PI-103 was used to attach a self-cyclizing molecule by an ethereal linkage to make the prodrug. The TDR cells have demonstrated high ROS levels as compared to their drug-sensitive counterparts using 2′,7′-dichlorofluorescin diacetate and MitoSOX assay [20]. In the presence of high level of oxidative stress in the tumor milieu, the prodrug undergoes self-cyclization to release PI-103, which inhibits the conversion of PIP2 to PIP3. This leads to blockade of downstream Akt [29]. Endogenously generated ROS has been demonstrated to bring about oxidation of PTEN, a tumor suppressor and results in sustained activation of the signaling pathway [23]. H2O2 also interacts with PTEN and oxidizes and inactivates the tumor suppressor by forming disulfide bonds between Cys124 residue and Cys71 in oxidized PTEN [43]. Utilizing the increase in physiological ROS levels could be an effective treatment to overcome resistance to BRAF and MEK inhibitors in BRAF-mutant melanoma.

RIDR-PI-103 in combination with ROS inducers such as doxorubicin or cytosine arabinoside demonstrated efficacy in vitro against acute myeloid leukemia cells. RIDR-PI-103 has reduced affinity for PI3K and withstands liver microsomes. RIDR-PI-103 undergoes oxidation and release PI-103 in the cells. The studies assessing the conversion of the prodrug to its active components were performed using HPLC and mass spectrometry [29]. In breast cancer cells RIDR-PI-103 in combination with doxorubicin inhibited cell proliferation, impaired PI3K signaling, and activated DNA damage [30]. Use of RIDR-PI-103 in cord blood cells and fibroblasts did not demonstrate toxicity [29,30]. Similarly in our study, we observed that RIDR-PI-103 had increased specificity in TDR cells over melanocytes at 5 µM (Figs. 1 and 2). Previous studies used ROS inducers including the chemotherapeutics doxorubicin or cytosine arabinoside to activate RIDR-PI-103. In our study, we did not utilize ROS inducer and still observed activation of the prodrug. We also observed the activity of RIDR-PI-103 at a concentration as low as 5 µM in the BRAF mutant-resistant cells. On the basis of our in-vitro data, we establish RIDR-PI-103 at 5 µM as the optimal concentration for treatment of BRAF-mutant resistant melanoma cells since 10 µM dose of RIDR-PI-103 was toxic to melanocytes. We still proceeded to test the 10 µM dose of RIDR-PI-103 on TDR cell lines to observe whether the higher concentration elicit effects similar to 5 µM of the prodrug. This bolsters our findings of the study that 10 µM RIDR-PI-103 used does not aberrantly re-activate the PI3K/Akt pathway in TDR cells but is toxic to cells. We observed some inconsistent data for WM983B TDR cell line in proliferation assays (Fig. 2) compared to cell viability assay using trypan blue (Fig. S3, Supplemental Digital Content 3, https://links.lww.com/ACD/A493). In Fig. 2, we are measuring the proliferation of TDR cells in the presence of RIDR-PI-103 using 2D and 3D assays, in Fig. S3, Supplemental Digital Content 3, https://links.lww.com/ACD/A493, we are measuring cell viability using the trypan blue exclusion assay. The variance in the results can be attributed to assay differences. For Fig. S3, Supplemental Digital Content 3, https://links.lww.com/ACD/A493, we have measured cell viability using trypan blue exclusion assay for up to 72 h. WM983B is derived from a metastatic melanoma whereas A375 and WM115 are primary melanomas [44–46]. The lack of effect of RIDR-PI-103 on cell viability using the trypan blue exclusion assay in WM983B TDR compared to other TDR cells could be attributed to the metastatic nature of WM983B cells and activation of other pro-survival signaling pathways in response to RIDR-PI-103 which antagonize the effect of this prodrug.

Our data show that RIDR-PI-103 inhibits its target, the PI3K/Akt pathway. We observed inhibition of p-AktS473 in WM115 TDR, A375 TDR, and WM983B TDR cells with 24 h of RIDR-PI-103 treatment in a dose-dependent manner. The inhibitory effect of RIDR-PI-103 via PI3K signaling was confirmed using a constitutively active AKT plasmid where transient transfection of WM115TDR cell using AKT constitutive active plasmid prior to RIDR-PI-103 treatment rescued the inhibitory effect induced by it. However, we could not confirm a similar mechanism in WM983TDR or A375TDR cell lines as they were not amendable to transient transfection. S6 is a protein that is downstream of Akt. Blockade of p-Akt using PI3K inhibitors also blocks p-S6 [47,48]. We observed blockade of p-S6 upon treatment with RIDR-PI-103 in a dose-dependent manner at residues Ser240/244 and Ser235/236 (Fig. 3). We did not observe attenuation in the mTOR1/2 pathway proteins in treatment with increasing concentrations of the prodrug (data not shown). In order to determine activation of RIDR-PI-103 relies on ROS levels, we treated cells with glutathione, ROS scavenger with or without RIDR-PI-103. Glutathione exists in a reduced and oxidized (GSSG) state in cells. Redox status of the cells is determined by the ratio of glutathione to GSSG. An excess of ROS is ultimately converted to H2O2 by the cells. In presence of H2O2, glutathione gets converted to GSSG and results in formation of water and oxygen, which in turn reduces oxidative stress in the cells [49,50]. When TDR cells were co-treated with glutathione, glutathione can quench ROS, and owing to decreases in ROS levels, the prodrug would have reduced activity. This was indicated by the rescue in cell proliferation that we observed in the groups treated with glutathione and RIDR-PI-103 compared to cells treated with RIDR-PI-103 alone (Fig. 4). Addition of NAC to RIDR-PI-103 rescued cell proliferation induced by RIDR-PI-103 in A375 TDR and WM983B TDR cell lines. Since NAC inhibited cell proliferation in the WM115 TDR cell line, interpretation of data for this cell line is difficult (Fig. S6, Supplemental Digital Content 6, https://links.lww.com/ACD/A496). Glutathione is composed of three amino acids – glycine, cysteine, and glutamic acid while NAC is the precursor for glutathione. A head-to-head study comparing the effects of NAC and glutathione on melanoma cells has not been conducted; we anticipate that glutathione might be a more stable molecule compared to NAC. A study examining glutathione levels in TDR cells will help to delineate if endogenous glutathione aids in the rescue of cell proliferation along with the addition of exogenous glutathione in these cell lines.

A corollary of using ROS scavenger was to use ROS inducer and assess the effect of ROS inducer in combination with RIDR-PI-103. When TDR cells are co-treated with RIDR-PI-103 and TBHP, the TBHP will increase ROS levels in the cells. More activation of RIDR-PI-103 is expected when treated with the combination opposed to RIDR-PI-103 alone which can be demonstrated by inhibition of cell proliferation. Combination of RIDR-PI-103 and TBHP significantly inhibited cell proliferation in WM115 TDR and WM983B TDR cells while we did not observe this for A375 TDR cell line (Fig. 5). This could possibly be attributed to differences in endogenous ROS levels in A375 TDR cells compared to WM115 TDR and WM983B TDR cells. A study simultaneously examining the ROS levels in all three TDR cell lines will provide further insights. Fig. 7 summarizes the efficacy of RIDR-PI-103 in BRAF and MEK inhibitor-resistant BRAF-mutant melanoma.

Fig. 7:

Summary of the efficacy of RIDR-PI-103 in BRAF and MEK inhibitor-resistant BRAF-mutant melanoma. RIDR-PI-103 (a prodrug of PI-103) is activated in a high ROS environment in TDR cells and releases biologically active PI-103. PI-103 is a pan-PI3K inhibitor and inhibits the activation of Akt and S6. RIDR-PI-103 inhibits cell proliferation, growth, and migration in TDR cells. Figure generated using biorender.com. ANOVA, analysis of variance; RIDR, ROS-induced drug release; ROS, reactive oxygen species.

Single-dose pharmacokinetics of RIDR-PI-103 revealed the half-life of this drug to be around 9 h based on a one-compartment model [30]. Our study has limitations. In-vivo studies using RIDR-PI-103 in resistant melanomas are beyond the scope of our current work and are a limitation of this study. It is important to assess the conversion of RIDR-PI-103 to PI-103 in vitro and in vivo. Since RIDR-PI-103 is a novel compound, there are challenges in terms of how to safely administer the prodrug and its efficacy in vivo. Pharmacokinetics studies focusing on single and repeated doses and dose escalation of RIDR-PI-103 will be carried out in the future. This will give a thorough understanding of the prodrug dosage and frequency. Future studies will be focused on developing the optimal dosage forms that can be utilized to administer the prodrug and circumvent toxicities and first-pass metabolism. In this study, we provide insight into the efficacy of RIDR-PI-103 in vitro in drug-resistant melanomas. Further studies investigating the safety and efficacy of RIDR-PI-103 are warranted on cancers resistant to BRAF and MEK inhibitors in BRAF-mutant melanoma.

Conclusion