Melanoma cells are highly resistant to anoikis, a form of apoptosis induced in nonadherent/inappropriate adhesion conditions. Depleting B-RAF or the prosurvival Bcl-2 family protein Mcl-1 renders mutant B-RAF melanoma cells susceptible to anoikis. In this study, we examined the effect of targeting B-RAF on the survival of primary stage melanoma cells cultured in three-dimensional type I collagen gels, which partially mimics the dermal microenvironment. Depletion/inhibition of B-RAF with small interfering RNA or the mutant B-RAF inhibitor, PLX4720, induced apoptosis of mutant B-RAF melanoma cells in three-dimensional collagen. Apoptosis was dependent on two upregulated BH3-only proteins, Bim-EL and Bmf, and was inhibited by ectopic Mcl-1 expression. Akt3 activation has been associated with the survival of melanoma cells. Mutant B-RAF melanoma cells ectopically expressing a constitutively activated form of Akt3 or endogenously expressing mutant Akt3 were protected from apoptosis induced by B-RAF knockdown or PLX4720 treatment. Furthermore, intrinsically resistant metastatic melanoma cells displayed elevated Akt phosphorylation in three-dimensional collagen and were rendered susceptible to PLX4720 by Akt3 knockdown. Importantly, myristylated Akt3 prevented B-RAF targeting–induced upregulation of Bim-EL and Bmf in three-dimensional collagen and partially protected Mcl-1–depleted cells from apoptosis. These findings delineate how mutant B-RAF protects melanoma cells from apoptosis and provide insight into possible resistance mechanisms to B-RAF inhibitors. Cancer Res; 70(16); 6670–81. ©2010 AACR.

Melanoma is the deadliest form of skin cancer, and its incidence is on the rise. It is treatable by surgical excision when detected at a noninvasive radial growth phase (RGP) stage. However, during the vertical growth phase (VGP), it invades through the basement membrane and into the dermis. Two major issues thwarting the efficacy of melanoma treatments are the propensity of VGP cells to metastasize from the dermis and resistance to chemotherapeutic regimens.

A significant advance in understanding the biology of melanoma was the identification of B-RAF mutations in ∼60% of these tumors (1). B-RAF is a serine-threonine kinase that activates the mitogen-activated protein/extracellular signal-regulated kinase (ERK) kinase (MEK)-ERK 1/2 pathway. The most frequent mutation in B-RAF is a valine to glutamate substitution at codon 600 (V600E), which results in B-RAF activation and occurs early in the disease. The downstream MEK-ERK1/2 pathway is activated in melanomas (2, 3) and is required for the proliferation, resistance to apoptosis, and invasion of mutant B-RAF melanoma cells (46). These findings have focused efforts on the development of inhibitors to target B-RAF in melanoma. Recently, a structure-guided approach led to the generation of PLX4720, a drug that binds near the ATP-binding site in B-RAF and exhibits a 10-fold lower IC50 toward B-RAFV600E compared with wild-type B-RAF (7). In an early-stage clinical trial, PLX4032, a structural analogue of PLX4720, showed strong antitumor effects in patients genotyped as harboring mutant B-RAF (8). However, the majority of patients administered with PLX4032 exhibit a partial response. In addition, B-RAF and MEK inhibitors in preclinical assays typically elicit a cytostatic response (4, 7, 9). These findings indicate that it will be critical to promote cytotoxic effects in B-RAF–targeted tumors to obtain prolonged tumor regression in patients.

A crucial step in tumor progression is the acquisition of resistance to anoikis, a form of apoptosis induced by loss of adhesion or inappropriate adhesion to the extracellular matrix (ECM; ref. 10). The ability of melanoma cells to counteract anoikis is critical as they move out from the epidermis, invade the type I collagen-rich dermis, survive in the circulatory system, and eventually establish metastatic colonies. We and others have shown that human melanocytes are susceptible to anoikis, whereas mutant B-RAF harboring VGP melanoma cells are resistant (11, 12). B-RAF signaling is implicated in the resistance of VGP melanoma cells because B-RAF depletion or MEK inhibition renders invasive cells susceptible to anoikis (12, 13). These effects were associated with enhanced expression of the proapoptotic BH3-only protein, B-cell lymphoma 2 interacting mediator extra long isoform (Bim-EL), and decreased levels of the prosurvival protein, myeloid cell leukemia-1 (Mcl-1; refs. 13, 14).

Integrin-mediated adhesion to the ECM component, fibronectin, protects B-RAF knockdown melanoma cells from undergoing apoptosis (12). Fibronectin is a minor component of the dermis (15) and is deposited at premetastatic niches (16). Furthermore, fibronectin binding integrins, αvβ3 and α4β1, are highly expressed in melanoma cells (17), and use of peptides to block cell binding to fibronectin prevents melanoma cell growth in mouse models (18). The mechanism of fibronectin-dependent protection from apoptosis is mediated, at least in part, through activation of Akt (12). Of the three Akt isoforms, Akt3 likely plays a key role because it is the predominately activated Akt isoform in melanoma cells, and knockdown of Akt3 cooperates with B-RAF knockdown to promote apoptosis (19, 20).

It will be critical to understand how to maximize the therapeutic effects of targeting B-RAF. To this end, we used inhibitor and RNA interference approaches to target B-RAF in mutant B-RAF melanoma cells. Analysis was done on RGP or VGP-like cells in three-dimensional collagen to more accurately recapitulate the dermal microenvironment. We show that targeting B-RAF induces apoptosis in three-dimensional collagen that is controlled by levels of the BH3-only proteins, Bim-EL and Bcl-2-modifying factor (Bmf), and the prosurvival protein Mcl-1. Constitutive activation of Akt3 protects against B-RAF knockdown/inhibition, prevents the upregulation of Bim-EL and Bmf, and partially protects melanoma cells from Mcl-1 depletion. Together, these findings delineate mutant B-RAF–dependent survival mechanisms in tumor cells and provide insight into possible modes of resistance to B-RAF inhibitors.

Cell culture

Human WM793, WM35, WM278, WM46, and 1205Lu cells were generously provided by Dr. Meenhard Herlyn (Wistar Institute, Philadelphia, PA) during the time period 2004 to 2009; A375 cells were purchased from the American Type Culture Collection. In all cases, early passage cultures were stored and used for the basis of these experiments. All cells were cultured in MCDB153 (Sigma-Aldrich) containing 20% Leibovitz L-15 medium, 2% fetal bovine serum (FBS) and 5 μg/mL insulin, except A375 cells that were cultured in DMEM with 10% FBS. WM793 and WM278 are VGP-like; WM35 display RGP properties; WM46, A375, and 1205Lu are metastatic lines.

Antibodies

The following antibodies were purchased from Cell Signaling Technology: anti-pan Akt (#9227), anti-Akt3 (#4059), anti–phospho-Akt (Ser473, #9271), anti–phospho-(Ser/Thr) Akt substrate (#9611), anti–phospho-ERK (Thr202/Tyr204, #4377), anti-HA (#2367), and anti–phospho-GSK3β (Ser9, #9336). Anti–Mcl-1 (#559027) and anti-GSK3β (#610201) were from BD Transduction. Anti-actin (#A2066) was from Sigma-Aldrich. Anti-Bim (#AAP-330) was from Stressgen.

Western blotting

Western blotting was done as previously described (12). Briefly, cells were washed and lysed directly in Laemmli sample buffer. Lysates were separated on SDS polyacrylamide gels and transferred to polyvinylidene difluoride membranes. Proteins were detected using the indicated primary antibody, and the signal was detected using peroxidase-conjugated secondary antibody followed by development using chemiluminescence substrate (Pierce). Chemiluminescence was detected using a Versadoc Imaging system (Bio-Rad) and was quantitated using the Quantity-One software.

Small interfering RNA transfections

WM793 cells were transfected with small interfering RNAs (siRNA) using Oligofectamine (Invitrogen) as previously described (12). Nontargeting control, B-RAF #1, Mcl-1 #11, Bmf #7, and Akt3 siRNAs were purchased from Dharmacon. Bim siRNA were from Cell Signaling Technology (13). All siRNAs were used at a final concentration of 25 nmol/L. Cells were transfected for 72 hours before subsequent treatment or analysis.

Generation of lentiviruses and cell lines

Wild-type Akt3 and NH2-terminal myristylated Akt3 were amplified from human Akt3 cDNA plasmid (Addgene) using the KOD Hot Start DNA polymerase kit (Novagen). The primer set for wild-type Akt3 was as follows: forward 5′ caccatgtacgcctacgacgtgcccg 3′ and reverse 5′ ttattctcgtccacttgcagagtagg 3′. Myr-HA-AKT3 was amplified in the same way except that forward primer 5′-caccatggggagcagcaagag-3′ was used in the PCR reaction. In both constructs, the HA epitope tag was retained. Akt fragments were then cloned into pENTR/D-TOPO vector (Invitrogen) according to the manufacturer's guidelines. Entry plasmids were recombined with pLenti6/UbC/V5-DEST using the LR Clonase II kit and protocol (Invitrogen) to generate pLenti6/UbC/HA-AKT3 and pLenti6/UbC/Myr-HA-AKT3. Constructs were verified by DNA sequencing. Lentiviruses were generated as described before (13) using the ViraPower Lentiviral Gateway Expression kit (Invitrogen). Cells were incubated for 72 hours in the presence of lentiviral supernatant. Transduced cell populations were selected with 5 μg/mL blasticidin.

Wild-type Mcl-1 cDNA sequence was PCR amplified from human Mcl-1 cDNA kindly provided by Dr. Steven Edwards (University of Liverpool, Liverpool, United Kingdom; ref. 21) using the KOD hot Start DNA polymerase kit and the primers 5′-caccatgtttggcctcaaaagaaccg-3′ and 5′-ctatcttattagatatgcc-3′. DNA fragment was cloned into pENTR/D-TOPO vector, and the entry plasmid was then recombined with pLenti4/TO/V5-DEST to form destination plasmid pLenti4/TO/Mcl-1. WM793 cells that stably express tet repressor (WM793TR; ref. 13) were infected with lentivirus for 72 hours before selection with 25 μg/mL zeocin.

Three-dimensional collagen gels

Collagen gels were cast by mixing the following on ice: Eagle's MEM (Lonza, Inc.), 2 mmol/L l-glutamine, 2% FBS, 0.15% sodium bicarbonate, and 0.8 mg/mL bovine type-I collagen (Organogenesis, Inc.). Cells were seeded in 2 mL collagen gels and incubated at 37°C for 30 minutes. After polymerization, the collagen gel lattice was overlaid with 2 mL medium.

Apoptosis assays

Three-dimensional collagen gels were dissolved in 1 mg/mL collagenase (Sigma-Aldrich) solution to release cells. Adherent cells in two-dimensional cell cultures were trypsinized. Cells were washed once in PBS and resuspended in 100 μL binding buffer (10 mmol/L HEPES, 140 mmol/L sodium chloride, 2.5 mmol/L calcium chloride) at a concentration of 1 × 106 cells/mL. Cells were then stained with 5 μL Annexin V-APC (BD Biosciences) for 15 minutes before 400 μL binding buffer was added. Staining was measured by flow cytometry on the FACSCalibur (BD Biosciences), and data were analyzed using Flowjo software (Three Star, Inc.).

Quantitative reverse transcription-PCR

Total RNA was extracted from melanoma cells by using the 5 Prime Perfect Pure RNA isolation kit (5 Prime, Inc.). RNA (1 μg) was reverse transcribed, and 1 of 20 of the resulting cDNA was used to detect mRNA abundance with primers for glyceraldehyde-3-phosphate dehydrogenase (forward, 5′-tggaccaccaactgcttag-3′; reverse, 5′-gatgcagggatgatgttc-3′), Bim-EL (forward, 5′-tccctgctgtctcgatcctc-3′; reverse, 5′-gctcttcggctgcttggtaa-3′), and Bmf (forward, 5′-gaggtacagattgcccgaaag-3′; reverse, 5′-ttcaaagcaaggttgtgca-3′). Reactions were done in SYBR Green mix and analyzed using the MyiQ real-time PCR detection system (Bio-Rad). Relative mRNA levels were calculated using the comparative Ct (ΔCt) method. Quantitation of mRNA levels represents data from three independent experiments.

Targeting B-RAF renders mutant B-RAF melanoma cells susceptible to apoptosis in three-dimensional collagen

We have previously shown that B-RAF knockdown renders the B-RAFV600E melanoma WM793 cells susceptible to apoptosis in suspension (12). Here, we tested the mutant-selective B-RAF inhibitor, PLX4720, in similar assays. The effectiveness of PLX4720 to inhibit B-RAF signaling in WM793 cells was manifested by its rapid and persistent inhibition of phosphoERK1/2 levels (Fig. 1A). Similar to our previous results with B-RAF knockdown, PLX4720 treatment increased Annexin V staining in suspended but not in fibronectin-adherent cells (Fig. 1B, quantitated in C). WM793 display properties of VGP, and to more faithfully recapitulate signaling in the dermal microenvironment, we extended our studies to three-dimensional type-I collagen gels. WM793 cells transfected with control siRNAs and subsequently cultured in three-dimensional collagen displayed low levels of Annexin V staining (Fig. 1D). By contrast, B-RAF knockdown or PLX4720-treated WM793 cells in three-dimensional collagen displayed enhanced Annexin V staining. We have previously observed similar results between Annexin V and cleaved caspase-3 staining, indicating activation of the intrinsic apoptotic pathway (14). These data show that mutant B-RAF is required for the protection of melanoma cells in suspension and three-dimensional collagen conditions. WM793 cells express collagen receptors including the α1 and α2 integrins (12), and α2β1 mediates cell arrest in fibrillar collagen gels in some melanoma lines (22). However, efficient knockdown of α1 and α2 integrins did not alter PLX4720-induced apoptosis in WM793 cells (Supplementary Fig. S1), indicating that these integrins are not required for the apoptosis mediated by B-RAF inhibition.

Figure 1.

PLX4720 treatment renders mutant B-RAF cells susceptible to apoptosis. A, WM793 cells were treated with 0.5 μmol/L PLX4720 for the times indicated and cell lysates analyzed by Western blotting with phosphoERK1/2 and total ERK2 antibodies. B, WM793 cells were plated on agar or fibronectin in the absence/presence of 0.5 μmol/L PLX4720. After 48 h, cells were harvested and stained with Annexin V-APC for apoptosis analysis by flow cytometry. X axis, fluorescence intensity; Y axis, cell counts, with percent of Annexin V-APC staining–positive cells indicated in each condition. C, quantitation of data from three independent experiments was represented by the mean percentage of cells staining positive for Annexin V-APC. D, WM793 cells were transfected with control or B-RAF siRNA. Seventy-two hours posttransfection, cells were harvested and seeded in three-dimensional collagen gels. Cells were treated with 0.5 μmol/L PLX4720 or DMSO (vehicle control). After 48 h, collagen gels were dissolved in collagenase solution, and collected cells were analyzed by Annexin V-APC staining and flow cytometry analysis.

Figure 1.

PLX4720 treatment renders mutant B-RAF cells susceptible to apoptosis. A, WM793 cells were treated with 0.5 μmol/L PLX4720 for the times indicated and cell lysates analyzed by Western blotting with phosphoERK1/2 and total ERK2 antibodies. B, WM793 cells were plated on agar or fibronectin in the absence/presence of 0.5 μmol/L PLX4720. After 48 h, cells were harvested and stained with Annexin V-APC for apoptosis analysis by flow cytometry. X axis, fluorescence intensity; Y axis, cell counts, with percent of Annexin V-APC staining–positive cells indicated in each condition. C, quantitation of data from three independent experiments was represented by the mean percentage of cells staining positive for Annexin V-APC. D, WM793 cells were transfected with control or B-RAF siRNA. Seventy-two hours posttransfection, cells were harvested and seeded in three-dimensional collagen gels. Cells were treated with 0.5 μmol/L PLX4720 or DMSO (vehicle control). After 48 h, collagen gels were dissolved in collagenase solution, and collected cells were analyzed by Annexin V-APC staining and flow cytometry analysis.

Close modal

Upregulation of Bim-EL and Bmf contributes to apoptosis in three-dimensional collagen following B-RAF targeting

BH3-only proteins, such as Bim-EL and Bmf, are upregulated following B-RAF knockdown or MEK inhibition in two-dimensional monolayer cultures of melanoma cells (13, 2325). Initially, we validated that the upregulation of Bim-EL and Bmf following B-RAF knockdown was efficiently prevented by siRNAs targeting either Bim or Bmf (Fig. 2A). Of note, we used quantitative reverse transcription-PCR (qRT-PCR) to detect Bmf due to the unavailability of a commercial antibody that detects a specific endogenous Bmf band, an observation consistent with Schmelzle and colleagues (26). Next, we addressed the contribution of Bim-EL and Bmf to apoptosis in B-RAF–depleted cells in three-dimensional collagen. Bim-EL depletion partially prevented apoptosis in B-RAF knockdown and PLX4720-treated WM793 cells in three-dimensional collagen (Fig. 2B). Similarly, Bmf depletion was effective at preventing apoptosis in B-RAF knockdown and PLX4720-treated cells. Furthermore, Bim-EL and Bmf coknockdown was more effective than either individual knockdown at preventing apoptosis in PLX4720-treated cells (Fig. 2B). These data show that both Bim-EL and Bmf are required for the apoptosis of B-RAF–targeted cells in three-dimensional collagen.

Figure 2.

Apoptosis in three-dimensional collagen following B-RAF targeting is regulated by Bcl-2 family proteins, Bim-EL, Bmf, and Mcl-1. A, WM793 cells were transfected with nontargeting (ctrl), B-RAF, Bim, and Bmf siRNAs, as indicated. After 72 h, cell lysates were analyzed by Western blotting for Bim-EL (left) or harvested for total RNA isolation and qRT-PCR analysis for Bmf (right). B, WM793 cells were transfected with control, B-RAF, Bim, and Bmf siRNAs, as indicated, for 72 h. Transfected cells were cultured in three-dimensional collagen gels for 48 h in the absence/presence of 0.5 μmol/L PLX4720 before analysis by flow cytometry for Annexin V. C, WM793 and WM793TR Mcl-1 cells were treated with or without 100 ng/mL doxycycline to regulate Mcl-1 expression and with or without PLX4720, as indicated. Cell lysates were analyzed by Western blotting for Mcl-1, phosphoERK1/2, and actin. D, WM793 and WM793TR Mcl-1 cells were treated with or without doxycycline for 3 d before being incorporated in three-dimensional collagen gels in serum-free conditions. Cells in three-dimensional collagen were treated with or without PLX4720 and, after 48 h, were processed for Annexin V staining. Columns, mean from three independent experiments; bars, SD.

Figure 2.

Apoptosis in three-dimensional collagen following B-RAF targeting is regulated by Bcl-2 family proteins, Bim-EL, Bmf, and Mcl-1. A, WM793 cells were transfected with nontargeting (ctrl), B-RAF, Bim, and Bmf siRNAs, as indicated. After 72 h, cell lysates were analyzed by Western blotting for Bim-EL (left) or harvested for total RNA isolation and qRT-PCR analysis for Bmf (right). B, WM793 cells were transfected with control, B-RAF, Bim, and Bmf siRNAs, as indicated, for 72 h. Transfected cells were cultured in three-dimensional collagen gels for 48 h in the absence/presence of 0.5 μmol/L PLX4720 before analysis by flow cytometry for Annexin V. C, WM793 and WM793TR Mcl-1 cells were treated with or without 100 ng/mL doxycycline to regulate Mcl-1 expression and with or without PLX4720, as indicated. Cell lysates were analyzed by Western blotting for Mcl-1, phosphoERK1/2, and actin. D, WM793 and WM793TR Mcl-1 cells were treated with or without doxycycline for 3 d before being incorporated in three-dimensional collagen gels in serum-free conditions. Cells in three-dimensional collagen were treated with or without PLX4720 and, after 48 h, were processed for Annexin V staining. Columns, mean from three independent experiments; bars, SD.

Close modal

Ectopic expression of Mcl-1 protects against PLX4720-induced apoptosis in three dimension

BH3-only proteins act, at least in part, by inhibiting prosurvival Bcl-2 proteins (27, 28). We have previously shown that B-RAF-MEK signaling enhanced protein stability of the prosurvival protein, Mcl-1 (14). To determine whether enhanced Mcl-1 expression was sufficient to confer resistance to apoptosis in B-RAF–targeted cells, we engineered an inducible Mcl-1 expression cell line. Following doxycycline treatment, enhanced expression of Mcl-1 was detected in nontreated and PLX4720-treated cells (Fig. 2C). Importantly, enhanced Mcl-1 expression protected WM793 cells from apoptosis induced by PLX4720 treatment in three-dimensional collagen (Fig. 2D). Mcl-1 also protected cells from apoptosis induced by MEK inhibition with U0126 (data not shown).

Constitutive activation of Akt3 protects against B-RAF inhibition

Adhesion to fibronectin, but not to collagen, enhances levels of phospho-Akt in WM793 cells (12), and Akt3 is the main Akt isoform activated in melanoma cells (19). Thus, we determined whether constitutive activation of Akt3 was able to protect B-RAF–targeted WM793 cells against apoptosis in three-dimensional collagen. Addition of the myristylation signal of the c-Src kinase to the NH2 terminus of Akt3 strongly enhances Akt3 activation and transforming activity in fibroblast focus–forming assays (29). We generated wild-type and myristylated human Akt3 lentiviruses and, subsequently, WM793 cell populations that stably express the Akt3 transgenes. Expression was confirmed by Western blotting with Akt and epitope-tag antibodies (Fig. 3A). Myristylated Akt3 but not wild-type Akt3 led to increased levels of phospho-Akt. Next, we determined the effect of adhesion on Akt signaling in the ectopic Akt-expressing cell populations. Although activation of endogenous Akt and wild-type Akt3 was adhesion dependent, myristylated Akt3 activity was constitutive and adhesion independent, as indicated by phospho-Akt staining (Fig. 3B). Furthermore, myristylated Akt3 enhanced phosphorylation of GSK-3β, an Akt substrate, and levels of a prominent staining band detected by a phospho-Akt substrate antibody. To determine the effect of constitutive Akt3 signaling on B-RAF–targeted cells, we knocked down B-RAF in Akt3-expressing cell lines. By Annexin V staining, expression of myristylated Akt3 partially protected B-RAF knockdown cells from apoptosis in three-dimensional collagen, whereas wild-type Akt3 was not protective (Fig. 3C). Similarly, expression of myristylated Akt3 but not wild-type Akt3 protected WM793 cells in three-dimensional collagen gels from PLX4720-induced apoptosis (Fig. 3D). Notably, expression of myristylated Akt3 did not enhance phospho-ERK1/2 in B-RAF knockdown cells (Supplementary Fig. S2).

Figure 3.

Expression of constitutively active Akt3 protects melanoma cells from B-RAF inhibition. A, whole-cell lysates from WM793, WM793 HA-Akt3, or WM793 Myr-HA-Akt3 cells were analyzed by Western blotting for HA, total Akt, and phosphoS473-Akt. Actin is the loading control. B, WM793, WM793 HA-Akt3, or WM793 Myr-HA-Akt3 cells were serum starved overnight and replated on agar or fibronectin-coated dishes for 1 h in serum-free medium. Cell lysates were analyzed by Western blotting for phospho-S473 Akt, total Akt, phospho GSK3β, GSK3β, and phospho Akt substrate. Actin was used as a loading control. C, WM793, WM793 HA-Akt3, or WM793 Myr-HA-Akt3 cells were transfected with either control or B-RAF siRNAs for 72 h. Cells were then seeded in three-dimensional collagen gels and cultured in serum-free medium. After 48 h, cells were analyzed by Annexin V staining. D, serum-starved WM793, WM793 HA-Akt3, or WM793 Myr-HA-Akt3 cells were seeded in three-dimensional collagen gels and cultured in serum-free medium containing DMSO or 0.5 μmol/L PLX4720. After 48 h, cells were analyzed as in C. Columns, mean from three independent experiments; bars, SD.

Figure 3.

Expression of constitutively active Akt3 protects melanoma cells from B-RAF inhibition. A, whole-cell lysates from WM793, WM793 HA-Akt3, or WM793 Myr-HA-Akt3 cells were analyzed by Western blotting for HA, total Akt, and phosphoS473-Akt. Actin is the loading control. B, WM793, WM793 HA-Akt3, or WM793 Myr-HA-Akt3 cells were serum starved overnight and replated on agar or fibronectin-coated dishes for 1 h in serum-free medium. Cell lysates were analyzed by Western blotting for phospho-S473 Akt, total Akt, phospho GSK3β, GSK3β, and phospho Akt substrate. Actin was used as a loading control. C, WM793, WM793 HA-Akt3, or WM793 Myr-HA-Akt3 cells were transfected with either control or B-RAF siRNAs for 72 h. Cells were then seeded in three-dimensional collagen gels and cultured in serum-free medium. After 48 h, cells were analyzed by Annexin V staining. D, serum-starved WM793, WM793 HA-Akt3, or WM793 Myr-HA-Akt3 cells were seeded in three-dimensional collagen gels and cultured in serum-free medium containing DMSO or 0.5 μmol/L PLX4720. After 48 h, cells were analyzed as in C. Columns, mean from three independent experiments; bars, SD.

Close modal

To determine the generality of these effects, we expressed myristylated Akt3 in other mutant B-RAF–harboring melanoma cells. WM35 and WM278 cells are B-RAFV600E expressing and representative of RGP and VGP, respectively. Both Bim-EL and Bmf were upregulated by PLX4720 in these cells (data not shown). Western blotting confirmed that myristylated Akt3 was highly active, and that PLX4720 inhibited ERK1/2 phosphorylation in WM35 cells (Fig. 4A). Importantly, expression of myristylated Akt3 protected WM35 cells from apoptosis induced by PLX4720 in three-dimensional collagen (Fig. 4B). Similarly, myristylated Akt3-transduced WM278 cells displayed enhanced phosphoAkt, and PLX4720 inhibited ERK1/2 in these cells (Fig. 4C). WM278 cells were more sensitive to PLX4720 treatment compared with WM793 cells, but myristylated Akt3-WM278 cells still exhibited resistance to PLX4720-induced apoptosis in three-dimensional collagen (Fig. 4D). Together, these data show that constitutive activation of Akt3 protects B-RAF–targeted primary stage melanoma cells from apoptosis.

Figure 4.

Myristylated Akt3 protects WM35 and WM278 cells from PLX4720-induced apoptosis in three dimensions. A, WM35 and WM35 Myr-HA-Akt3 were treated with DMSO or 1 μmol/L PLX4720 for 1 h. Cell lysates were analyzed by Western blotting for phosphoS473 Akt, total Akt, phospho ERK1/2, and total ERK1/2. B, serum-starved WM35 and WM35 Myr-HA-Akt3 cells were seeded in three-dimensional collagen gels and cultured in serum-free medium containing DMSO or 1 μmol/L PLX4720. After 48 h, cells were analyzed by Annexin V staining. Columns, mean from three independent experiments; bars, SD. C, as for A except that WM278 and WM278 Myr-HA-Akt3 were used. D, as for B except that WM278 and WM278 Myr-HA-Akt3 were used, and cells were harvested 24 h following PLX4720 treatment.

Figure 4.

Myristylated Akt3 protects WM35 and WM278 cells from PLX4720-induced apoptosis in three dimensions. A, WM35 and WM35 Myr-HA-Akt3 were treated with DMSO or 1 μmol/L PLX4720 for 1 h. Cell lysates were analyzed by Western blotting for phosphoS473 Akt, total Akt, phospho ERK1/2, and total ERK1/2. B, serum-starved WM35 and WM35 Myr-HA-Akt3 cells were seeded in three-dimensional collagen gels and cultured in serum-free medium containing DMSO or 1 μmol/L PLX4720. After 48 h, cells were analyzed by Annexin V staining. Columns, mean from three independent experiments; bars, SD. C, as for A except that WM278 and WM278 Myr-HA-Akt3 were used. D, as for B except that WM278 and WM278 Myr-HA-Akt3 were used, and cells were harvested 24 h following PLX4720 treatment.

Close modal

PLX4720-resistant cell lines display elevated Akt phosphorylation in three-dimensional collagen

A small percentage of melanoma lines harbor Akt3 mutations (30). One identified Akt3 mutant line is WM46, which harbors an E17K mutation that enhances Akt association with the plasma membrane and increases Akt kinase activity (31). WM46 displayed enhanced Akt phosphorylation in three-dimensional collagen and resistance to PLX4720-induced apoptosis (Fig. 5A). Even treatment with high concentrations of PLX4720 (5–10 μmol/L) were unable to promote substantial apoptosis in WM46 cells. During our studies, we identified two metastatic lines, A375 and 1205Lu, which were de novo resistant to PLX4720-induced apoptosis in three-dimensional collagen (Fig. 5B). Both A375 and 1205Lu displayed enhanced Akt phosphorylation in three dimension compared with the sensitive lines WM35 and WM278. Furthermore, knockdown of Akt3 rendered 1205Lu cells sensitive to PLX4720-induced apoptosis in three dimensions (Fig. 5C). These data further associate elevated Akt phosphorylation with resistance to PLX4720.

Figure 5.

Akt activity in three-dimensions is elevated in PLX4720-resistant cell lines. A, WM278 and WM46 cells were seeded into three-dimensional collagen gels with or without PLX4720 (1, 5, and 10 μmol/L, as indicated). Cells were lysed after 24 h for Western blot analysis for phosphoS473 Akt, phospho ERK1/2, and actin (left). WM46 cells were analyzed after 48 h for Annexin V staining (right). B, A375 and 1205Lu cells were seeded into three-dimensional collagen gels. Left, Annexin V staining following 48-h treatment with 5 μmol/L PLX4720. Right, Western blot analysis for phosphoS473 Akt, phospho ERK1/2, and actin in A375 and 1205Lu cells in three-dimensions compared with WM35 and WM278. C, 1205Lu cells were transfected with control nontargeting or Akt3 siRNAs for 72 h. Cells were lysed for Western blot analysis for phosphoS473 Akt, Akt3, and actin (left). Cells were seeded in three-dimensional collagen gels with or without 5 μmol/L PLX4720. After 48 h, cells were harvested for Annexin V staining and flow cytometry analysis (right).

Figure 5.

Akt activity in three-dimensions is elevated in PLX4720-resistant cell lines. A, WM278 and WM46 cells were seeded into three-dimensional collagen gels with or without PLX4720 (1, 5, and 10 μmol/L, as indicated). Cells were lysed after 24 h for Western blot analysis for phosphoS473 Akt, phospho ERK1/2, and actin (left). WM46 cells were analyzed after 48 h for Annexin V staining (right). B, A375 and 1205Lu cells were seeded into three-dimensional collagen gels. Left, Annexin V staining following 48-h treatment with 5 μmol/L PLX4720. Right, Western blot analysis for phosphoS473 Akt, phospho ERK1/2, and actin in A375 and 1205Lu cells in three-dimensions compared with WM35 and WM278. C, 1205Lu cells were transfected with control nontargeting or Akt3 siRNAs for 72 h. Cells were lysed for Western blot analysis for phosphoS473 Akt, Akt3, and actin (left). Cells were seeded in three-dimensional collagen gels with or without 5 μmol/L PLX4720. After 48 h, cells were harvested for Annexin V staining and flow cytometry analysis (right).

Close modal

Myristylated Akt3 prevents upregulation of Bim-EL and Bmf mRNA following B-RAF inhibition

Next, we examined the effect of Akt3 signaling on the induction of Bim-EL and Bmf following B-RAF knockdown and PLX4720 treatment in three-dimensional collagen. Total RNA was extracted from WM793 cells in collagen gels and analyzed by qRT-PCR. As expected, Bim-EL and Bmf mRNA levels were increased in three dimensions following B-RAF knockdown (Fig. 6A and B). In cells expressing myristylated Akt3, the upregulation of both Bim-EL and Bmf was inhibited. Similarly, treatment with PLX4720 in WM793 cells in three dimensions led to the upregulation of Bim-EL and Bmf, which was inhibited in cells expressing myristylated Akt3 (Fig. 6A and B). Thus, the protective effect of myristylated Akt3 in B-RAF–targeted cells is associated with an impaired upregulation of Bim-EL and Bmf mRNA.

Figure 6.

Myristylated Akt3 counteracts the upregulation of BH3-only proteins and protects against Mcl-1 knockdown–induced apoptosis in three dimensions. A, WM793 and WM793 Myr-HA-Akt3 cells were transfected with either control or B-RAF siRNAs for 72 h. Cells were then seeded in three-dimensional collagen gels and cultured in serum-free medium. After 24 h, cells were harvested for total RNA isolation and qRT-PCR analysis. Additionally, WM793 and WM793 Myr-HA-Akt3 cells were treated with 0.5 μmol/L PLX4720 for 40 h before harvesting cells. Quantitation of data from three independent experiments is represented as the mean relative mRNA level of Bim-EL in each condition. B, as for A, except that Bmf mRNA levels were analyzed. C, WM793, WM793 HA-AKT3, or WM793 Myr-HA-AKT3 cells were transfected with control or Mcl-1 siRNA. Seventy-two hours posttransfection, cell lysates were harvested and analyzed by Western blotting for Mcl-1 and actin. D, WM793, WM793 HA-AKT3, or WM793 Myr-HA-AKT3 cells transfected with control or Mcl-1 siRNAs were cultured in three-dimensional collagen gels in serum-free conditions. After 48 h, cells were harvested and stained with Annexin V-APC for flow cytometry. Columns, mean from four independent experiments; bars, SD.

Figure 6.

Myristylated Akt3 counteracts the upregulation of BH3-only proteins and protects against Mcl-1 knockdown–induced apoptosis in three dimensions. A, WM793 and WM793 Myr-HA-Akt3 cells were transfected with either control or B-RAF siRNAs for 72 h. Cells were then seeded in three-dimensional collagen gels and cultured in serum-free medium. After 24 h, cells were harvested for total RNA isolation and qRT-PCR analysis. Additionally, WM793 and WM793 Myr-HA-Akt3 cells were treated with 0.5 μmol/L PLX4720 for 40 h before harvesting cells. Quantitation of data from three independent experiments is represented as the mean relative mRNA level of Bim-EL in each condition. B, as for A, except that Bmf mRNA levels were analyzed. C, WM793, WM793 HA-AKT3, or WM793 Myr-HA-AKT3 cells were transfected with control or Mcl-1 siRNA. Seventy-two hours posttransfection, cell lysates were harvested and analyzed by Western blotting for Mcl-1 and actin. D, WM793, WM793 HA-AKT3, or WM793 Myr-HA-AKT3 cells transfected with control or Mcl-1 siRNAs were cultured in three-dimensional collagen gels in serum-free conditions. After 48 h, cells were harvested and stained with Annexin V-APC for flow cytometry. Columns, mean from four independent experiments; bars, SD.

Close modal

Activated Akt3 promotes cell survival in Mcl-1–depleted melanoma cells in three dimension

We have previously shown that downregulation of Mcl-1 renders melanoma cells susceptible to apoptosis in suspension (14). Extending these observations to three dimensions, efficient downregulation of Mcl-1 promoted apoptosis in collagen gels (Fig. 6C and D). If Akt3 protective effects were mediated through Bim-EL and Bmf, we would predict that myristylated Akt3 would elicit protective effects in Mcl-1 knockdown cells. Expression of myristylated Akt3 but not wild-type Akt3 was sufficient to protect Mcl-1 knockdown cells from apoptosis (Fig. 6D). Notably, the protection was not as complete as effects elicited by myristylated Akt3 in PLX4720-treated cells; nevertheless, these data suggest that Akt3 signaling regulates apoptotic signaling events, at least in part, through Bim-EL and Bmf.

The identification of B-RAF mutations in melanoma, and to a lesser extent in thyroid carcinoma, colorectal cancer, and ovarian cancer, has focused efforts on targeted therapeutic strategies to treat these malignancies. Here, we have used inhibitor and RNA interference approaches to determine the effect of targeting B-RAF in primary stage melanoma cells cultured in three-dimensional dermal-like matrices. It is critical to test the importance of tumor cell signaling pathways in ECM microenvironments that more closely resemble the in vivo situation compared with two-dimensional cultures. Other groups have shown that Rho GTPase activity and focal adhesion kinase signaling differ between three-dimensional and two-dimensional conditions (32, 33). Furthermore, the three-dimensional environment also controls cellular differentiation (34) and has been used in breast and prostate cancer progression models (35, 36).

Initially, our study shows that either B-RAF depletion or treatment with the recently described mutant-selective B-RAF inhibitor (7), PLX4720, promotes apoptosis in three-dimensional collagen gels. The effects that we observe are similar to the susceptibility of B-RAF–targeted cells to apoptosis in suspension conditions (13). Although primary stage cell lines used in this study are sensitive to PLX4720-induced apoptosis, some metastatic lines displayed elevated Akt phosphorylation in three-dimensional and were less susceptible. Addition of fibronectin to three-dimensional collagen gels elicited partial protection to PLX4720-induced apoptosis in WM35 cells, and low level of protection was observed in two other susceptible cell lines (data not shown). The effects may not be as dramatic as in two-dimension due to inefficient remodeling of fibronectin or lower stiffness of the three-dimensional system.

Enhanced apoptosis in B-RAF–targeted cells in three dimensions was dependent on the BH3-only proteins Bim-EL and Bmf. We and others have previously shown that Bim-EL is upregulated following B-RAF knockdown or MEK inhibition in melanoma cells and sensitizes melanoma cells to anoikis (13, 23, 24). Upregulation of Bim-EL occurs through mRNA regulation and protein stability mechanisms (13). Others have suggested that cytosolic accumulation of Bmf correlates with apoptotic sensitivity to MEK inhibition (25); however, the effect of altered Bmf expression in melanoma cells remains unclear. In our studies, Bmf mRNA levels were upregulated following B-RAF knockdown. These findings are consistent with recent microarray analyses (37). We were unable to verify the specificity of commercial Bmf antibodies, similar to concerns raised by other laboratories (26). Although both Bim-EL and Bmf mRNA levels are upregulated by loss of adhesion in epithelial and fibroblast cells (26, 38, 39), Bim-EL and Bmf mRNA levels were not adhesion regulated in mutant B-RAF melanoma cells (data not shown). Bmf was originally identified in a screen as a binding partner for Mcl-1 (40), and ERK1/2 activity regulates Mcl-1 protein stability in melanoma cells (14). We show that enhanced expression of Mcl-1 provides melanoma cells with resistance to PLX4720-induced apoptosis. Mcl-1 expression in melanoma may underlie resistance to several distinct types of proapoptotic signals including resistance to dacarbazine treatment (41), exposure to ionizing radiation (42), the proteasome inhibitor Bortezomib (43), and endoplasmic reticulum stress (44). These findings suggest that BH3 mimetics that target Mcl-1 may prove therapeutically useful in the treatment of melanoma.

Adhesion of melanoma cells to fibronectin but not type I collagen activates Akt, an effect observed even in PTEN-deficient cells such as WM793 (12). Akt3 is the main Akt isoform activated in melanoma (19), and we show that ectopically expressed myristylated Akt3 signals in an adhesion-independent manner. Furthermore, constitutive Akt3 activity rendered melanoma cells resistant to B-RAF targeting through either B-RAF knockdown or PLX4720 treatment. Akt3 actions are, at least in part, mediated through preventing the upregulation of Bim-EL and Bmf induced by B-RAF inhibition. Others have implicated activated Akt in the suppression of Bim-EL and Bmf in nonmelanoma cell types (26, 45), and these findings indicate that other Akt isoforms may elicit similar effects. We have not yet identified the downstream effectors of Akt3 that mediate effects on Bim-EL and Bmf; however, based on others findings mammalian target of rapamycin, PRAS40 and GSK-3β are likely mediators (4648). Similar to B-RAF knockdown/inhibition, depletion of the prosurvival protein Mcl-1 renders melanoma cells susceptible to apoptosis in three dimension. Mcl-1 knockdown–induced apoptosis was protected by the expression of myristylated Akt3; however, the effect was partial, suggesting that additional pathways may be involved.

In summary, we show that constitutive, adhesion-independent Akt3 activity confers resistance to melanoma cell apoptosis induced by B-RAF targeting in three dimensions. Our findings underscore the importance of combinatorial targeting of both the B-RAF-MEK and integrin-PI-3 kinase-Akt signaling pathways to elicit cytotoxic effects. Dual targeting of these two pathways inhibits melanoma tumors in xenografts and causes tumor regression of B-RAFV600E/loss of PTEN-induced melanomas in mouse models (20, 46, 49). Furthermore, overexpression of activated Akt1 enhances the tumor-forming ability of the nontumorigenic cell line WM35 (50). A small percentage of melanoma cells harbor Akt3 mutations (30); nevertheless, mutations in Akt3 or other components of this pathway represent possible mechanisms of de novo and/or acquired resistance to B-RAF inhibitors.

No potential conflicts of interest were disclosed.

We thank Dr. Gideon Bollag (Plexxikon) for providing PLX4720, Dr. Meenhard Herlyn (Wistar Institute) for the cell lines, Dr. Steven Edwards (Liverpool, United Kingdom) for providing the Mcl-1 cDNA, and the Flow Cytometry shared resource in the Kimmel Cancer Center.

Grant Support: NIH grants R01-GM067893 and R01-CA125103, and American Cancer Society grant RSG-08-03-01-CSM (A.E. Aplin).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Davies
H
,
Bignell
GR
,
Cox
C
, et al
. 
Mutations of the BRAF gene in human cancer
.
Nature
2002
;
417
:
949
54
.
2
Cohen
C
,
Zavala-Pompa
A
,
Sequeira
JH
, et al
. 
Mitogen-actived protein kinase activation is an early event in melanoma progression
.
Clin Cancer Res
2002
;
8
:
3728
33
.
3
Satyamoorthy
K
,
Li
G
,
Gerrero
MR
, et al
. 
Constitutive mitogen-activated protein kinase activation in melanoma is mediated by both BRAF mutations and autocrine growth factor stimulation
.
Cancer Res
2003
;
63
:
756
9
.
4
Hingorani
SR
,
Jacobetz
MA
,
Robertson
GP
,
Herlyn
M
,
Tuveson
DA
. 
Suppression of BRAF(V599E) in human melanoma abrogates transformation
.
Cancer Res
2003
;
63
:
5198
202
.
5
Karasarides
M
,
Chiloeches
A
,
Hayward
R
, et al
. 
B-RAF is a therapeutic target in melanoma
.
Oncogene
2004
;
23
:
6292
8
.
6
Bhatt
KV
,
Spofford
LS
,
Aram
G
,
McMullen
M
,
Pumiglia
K
,
Aplin
AE
. 
Adhesion control of cyclin D1 and p27Kip1 levels is deregulated in melanoma cells through BRAF-MEK-ERK signaling
.
Oncogene
2005
;
12
:
3459
71
.
7
Tsai
J
,
Lee
JT
,
Wang
W
, et al
. 
Discovery of a selective inhibitor of oncogenic B-Raf kinase with potent antimelanoma activity
.
Proc Natl Acad Sci U S A
2008
;
105
:
3041
6
.
8
Sondak
VK
,
Smalley
K
. 
Targeting mutant BRAF and KIT in metastatic melanoma: ASCO 2009 meeting report
.
Pigment Cell Melanoma Res
2009
;
22
:
386
7
.
9
Wellbrock
C
,
Ogilvie
L
,
Hedley
D
, et al
. 
V599EB-RAF is an oncogene in melanocytes
.
Cancer Res
2004
;
64
:
2338
42
.
10
Frisch
SM
,
Francis
H
. 
Disruption of epithelial cell-matrix interactions induces apoptosis
.
J Cell Biol
1994
;
124
:
619
26
.
11
Scott
G
,
Ryan
DH
,
McCarthy
JB
. 
Molecular mechanisms of human melanocyte attachment to fibronectin
.
J Invest Dermatol
1992
;
99
:
787
94
.
12
Boisvert-Adamo
K
,
Aplin
AE
. 
B-RAF and PI-3 kinase signaling protect melanoma cells from anoikis
.
Oncogene
2006
;
25
:
4848
56
.
13
Boisvert-Adamo
K
,
Aplin
AE
. 
Mutant B-RAF mediates resistance to anoikis via Bad and Bim
.
Oncogene
2008
;
27
:
3301
12
.
14
Boisvert-Adamo
K
,
Longmate
W
,
Abel
EV
,
Aplin
AE
. 
Mcl-1 is required for melanoma resistance to anoikis
.
Mol Cancer Res
2009
;
7
:
549
56
.
15
Tomasek
JJ
,
Gabbiani
G
,
Hinz
B
,
Chaponnier
C
,
Brown
RA
. 
Myofibroblasts and mechano-regulation of connective tissue remodelling
.
Nat Rev Mol Cell Biol
2002
;
3
:
349
63
.
16
Kaplan
RN
,
Riba
RD
,
Zacharoulis
S
, et al
. 
VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche
.
Nature
2005
;
438
:
820
7
.
17
Albelda
SM
,
Mette
SA
,
Elder
DE
, et al
. 
Integrin distribution in malignant melanoma: association of the β 3 subunit with tumor progression
.
Cancer Res
1990
;
50
:
6757
64
.
18
Humphries
MJ
,
Olden
K
,
Yamada
KM
. 
A synthetic peptide from fibronectin inhibits experimental metastasis of murine melanoma cells
.
Science
1986
;
233
:
467
70
.
19
Stahl
JM
,
Sharma
A
,
Cheung
M
, et al
. 
Deregulated Akt3 activity promotes development of malignant melanoma
.
Cancer Res
2004
;
64
:
7002
10
.
20
Cheung
M
,
Sharma
A
,
Madhunapantula
SV
,
Robertson
GP
. 
Akt3 and mutant V600EB-Raf cooperate to promote early melanoma development
.
Cancer Res
2008
;
68
:
3429
39
.
21
Akgul
C
,
Moulding
DA
,
White
MRH
,
Edwards
SW
. 
In vivo localisation and stability of human Mcl-1 using green fluorescent protein (GFP) fusion proteins
.
FEBS Lett
2000
;
478
:
72
6
.
22
Henriet
P
,
Zhong
ZD
,
Brooks
PC
,
Weinberg
KI
,
DeClerck
YA
. 
Contact with fibrillar collagen inhibits melanoma cell proliferation by up-regulating p27KIP1
.
Proc Natl Acad Sci U S A
2000
;
97
:
10026
31
.
23
Verhaegen
M
,
Bauer
JA
,
Martin de la Vega
C
, et al
. 
A novel BH3 mimetic reveals a mitogen-activated protein kinase-dependent mechanism of melanoma cell death controlled by p53 and reactive oxygen species
.
Cancer Res
2006
;
66
:
11348
59
.
24
Cartlidge
RA
,
Thomas
GR
,
Cagnol
S
, et al
. 
Oncogenic BRAF(V600E) inhibits BIM expression to promote melanoma cell survival
.
Pigment Cell Melanoma Res
2008
;
21
:
534
44
.
25
VanBrocklin
MW
,
Verhaegen
M
,
Soengas
MS
,
Holmen
SL
. 
Mitogen-activated protein kinase inhibition induces translocation of Bmf to promote apoptosis in melanoma
.
Cancer Res
2009
;
69
:
1985
94
.
26
Schmelzle
T
,
Mailleux
AA
,
Overholtzer
M
, et al
. 
Functional role and oncogene-regulated expression of the BH3-only factor Bmf in mammary epithelial anoikis and morphogenesis
.
Proc Natl Acad Sci U S A
2007
;
104
:
3787
92
.
27
Youle
RJ
,
Strasser
A
. 
The BCL-2 protein family: opposing activities that mediate cell death
.
Nat Rev Mol Cell Biol
2008
;
9
:
47
59
.
28
Merino
D
,
Giam
M
,
Hughes
PD
, et al
. 
The role of BH3-only protein Bim extends beyond inhibiting Bcl-2-like prosurvival proteins
.
J Cell Biol
2009
;
186
:
355
62
.
29
Mende
I
,
Malstrom
S
,
Tsichlis
P
,
Vogt
P
,
Aoki
M
. 
Oncogenic transformation induced by membrane-targeted Akt2 and Akt3
.
Oncogene
2001
;
20
:
4419
23
.
30
Davies
MA
,
Stemke-Hale
K
,
Tellez
C
, et al
. 
A novel AKT3 mutation in melanoma tumours and cell lines
.
Br J Cancer
2008
;
99
:
1265
8
.
31
Carpten
JD
,
Faber
AL
,
Horn
C
, et al
. 
A transforming mutation in the pleckstrin homology domain of AKT1 in cancer
.
Nature
2007
;
448
:
439
44
.
32
Cukierman
E
,
Pankov
R
,
Stevens
DR
,
Yamada
KM
. 
Taking cell-matrix adhesions to the third dimension
.
Science
2001
;
294
:
1708
12
.
33
Wozniak
MA
,
Desai
R
,
Solski
PA
,
Der
CJ
,
Keely
PJ
. 
ROCK-generated contractility regulates breast epithelial cell differentiation in response to the physical properties of a three-dimensional collagen matrix
.
J Cell Biol
2003
;
163
:
583
95
.
34
Schmeichel
KL
,
Bissell
MJ
. 
Modeling tissue-specific signaling and organ function in three dimensions
.
J Cell Sci
2003
;
116
:
2377
88
.
35
Petersen
O
,
Ronnov-Jessen
L
,
Howlett
A
,
Bissell
M
. 
Interaction with basement membrane serves to rapidly distinguish growth and differentiation pattern of normal and malignant human breast epithelial cells
.
Proc Natl Acad Sci U S A
1992
;
89
:
9064
8
.
36
Lang
SH
,
Sharrard
RM
,
Stark
M
,
Villette
JM
,
Maitland
NJ
. 
Prostate epithelial cell lines form spheroids with evidence of glandular differentiation in three-dimensional Matrigel cultures
.
Br J Cancer
2001
;
85
:
590
9
.
37
Packer
LM
,
East
P
,
Reis-Filho
JS
,
Marais
R
. 
Identification of direct transcriptional targets of V600E BRAF/MEK signalling in melanoma
.
Pigment Cell Melanoma Res
2009
;
22
:
785
98
.
38
Reginato
MJ
,
Mills
KR
,
Paulus
JK
, et al
. 
Integrins and EGFR coordinately regulate the pro-apoptotic protein Bim to prevent anoikis
.
Nat Cell Biol
2003
;
6
:
6
.
39
Woods
NT
,
Yamaguchi
H
,
Lee
FY
,
Bhalla
KN
,
Wang
H-G
. 
Anoikis, initiated by Mcl-1 degradation and Bim induction, is deregulated during oncogenesis
.
Cancer Res
2007
;
67
:
10744
52
.
40
Puthalakath
H
,
Villunger
A
,
O'Reilly
LA
, et al
. 
Bmf: A proapoptotic BH3-only protein regulated by interaction with the myosin V actin motor complex, activated by anoikis
.
Science
2001
;
293
:
1829
32
.
41
Thallinger
C
,
Wolschek
MF
,
Wacheck
V
, et al
. 
Mcl-1 antisense therapy chemosensitizes human melanoma in a SCID mouse xenotransplantation model
.
J Invest Dermatol
2003
;
120
:
1081
6
.
42
Skvara
H
,
Thallinger
C
,
Wacheck
V
, et al
. 
Mcl-1 blocks radiation-induced apoptosis and inhibits clonogenic cell death
.
Anticancer Res
2005
;
25
:
2697
703
.
43
Qin
JZ
,
Xin
H
,
Sitailo
LA
,
Denning
MF
,
Nickoloff
BJ
. 
Enhanced killing of melanoma cells by simultaneously targeting Mcl-1 and NOXA
.
Cancer Res
2006
;
66
:
9636
45
.
44
Jiang
CC
,
Lucas
K
,
Avery-Kiejda
KA
, et al
. 
Up-regulation of Mcl-1 is critical for survival of human melanoma cells upon endoplasmic reticulum stress
.
Cancer Res
2008
;
68
:
6708
17
.
45
Stahl
M
,
Dijkers
PF
,
Kops
GJPL
, et al
. 
The forkhead transcription factor FoxO regulates transcription of p27Kip1 and Bim in response to IL-2
.
J Immunol
2002
;
168
:
5024
31
.
46
Dankort
D
,
Curley
DP
,
Cartlidge
RA
, et al
. 
Braf(V600E) cooperates with Pten loss to induce metastatic melanoma
.
Nat Genet
2009
;
41
:
544
52
.
47
Madhunapantula
SV
,
Sharma
A
,
Robertson
GP
. 
PRAS40 deregulates apoptosis in malignant melanoma
.
Cancer Res
2007
;
67
:
3626
36
.
48
Panka
DJ
,
Cho
DC
,
Atkins
MB
,
Mier
JW
. 
GSK-3 inhibition enhances Sorafenib-induced apoptosis in melanoma cell lines
.
J Biol Chem
2008
;
283
:
726
32
.
49
Tran
MA
,
Gowda
R
,
Sharma
A
, et al
. 
Targeting V600EB-Raf and Akt3 using nanoliposomal-small interfering RNA inhibits cutaneous melanocytic lesion development
.
Cancer Res
2008
;
68
:
7638
49
.
50
Govindarajan
B
,
Sligh
JE
,
Vincent
BJ
, et al
. 
Overexpression of Akt converts radial growth melanoma to vertical growth melanoma
.
J Clin Invest
2007
;
117
:
719
29
.

Supplementary data