Isolation of a Novel Thioflavin S–Derived Compound That Inhibits

Published OnlineFirst September 18, 2013; DOI: 10.1158/1535-7163.MCT-13-0142

Molecular Cancer Small Molecule Therapeutics Therapeutics

Isolation of a Novel Thioﬂavin S–Derived Compound That Inhibits BAG-1–Mediated Protein Interactions and Targets BRAF Inhibitor–Resistant Cell Lines

Marion Enthammer1, Emmanouil S. Papadakis3, Maria Salome Gachet2, Martin Deutsch1, Stefan Schwaiger2, Katarzyna Koziel1, Muhammad Imtiaz Ashraf1, Sana Khalid1, Gerhard Wolber2, Graham Packham3, Ramsey I. Cutress3, Hermann Stuppner2, and Jakob Troppmair1

Abstract Protein–protein interactions mediated through the C-terminal Bcl-2–associated athanogene (BAG) domain of BAG-1 are critical for cell survival and proliferation. Thioflavin S (NSC71948)—a mixture of compounds resulting from the methylation and sulfonation of primulin base—has been shown to dose- dependently inhibit the interaction between BAG-1 and Hsc70 in vitro. In human breast cancer cell lines, with high BAG-1 expression levels, Thioflavin S reduces the binding of BAG-1 to Hsc70, Hsp70, or CRAF and decreases proliferation and viability. Here, we report the development of a protocol for the purification and isolation of biologically active constituents of Thioflavin S and the characterization of the novel compound Thio-2. Thio-2 blocked the growth of several transformed cell lines, but had much weaker effects on untransformed cells. Thio-2 also inhibited the proliferation of melanoma cell lines that had become resistant to treatment with PLX4032, an inhibitor of mutant BRAF. In transformed cells, Thio-2 interfered with intracellular signaling at the level of RAF, but had no effect on the activation of AKT. Thio-2 decreased binding of BAG-1 to Hsc70 and to a lesser extent BRAF in vitro and in vivo, suggesting a possible mechanism of action. Given that tumors frequently develop resistance to kinase inhibitors during treatment, Thio-2 and related compounds may offer promising alternative strategies to currently available therapies. Mol Cancer Ther; 12(11); 2400–14. 2013 AACR.

Introduction ylation and requires the assembly of signalosomes, which, RAF kinases are part of an evolutionarily conserved along with the core components of the signaling cascade, core-signaling cascade downstream of activated receptor contain accessory molecules required for modulation, tyrosine kinases and the small G-protein RAS (1). The compartmentalization, and specificity (1–3). RAF kinases three RAF isoforms, ARAF, BRAF, and CRAF (also RAF-1), are upstream of a three-tiered mitogen–activated protein are involved in mitogen, survival, and differentiation kinase (MAPK) cascade comprising the dual-specificity signaling. Their activation is a complex process involving Ser/Thr MAPK kinases (MAP2K) MEK1/2 and their interactions with proteins and lipids as well as phosphor- substrates ERK1/2, which are required for most reported RAF effects (4, 5). Several proteins have been described, which may be important for RAF activation and signaling under specific 1 Authors' Affiliations: Daniel Swarovski Research Laboratory, Depart- conditions at distinct cellular sites. These include the Bcl- ment of Visceral, Transplant and Thoracic Surgery, Innsbruck Medical University; 2Institute of Pharmacy/Pharmacognosy, Center of Molecular 2–associated athanogene 1 (BAG-1; refs. 6–10). BAG-1 was Biosciences, University of Innsbruck, Innsbruck, Austria; and 3Cancer originally identified as a Bcl-2–interacting protein with Research UK Centre, Cancer Sciences Division, University of Southampton Faculty of Medicine, Southampton General Hospital, Southampton, United antiapoptotic activity (11). BAG-1 is a member of a family Kingdom of proteins characterized by at least one copy of an Note: Supplementary data for this article are available at Molecular Cancer approximately 100 amino acid–long evolutionarily con- Therapeutics Online (http://mct.aacrjournals.org/). served a-helical BAG domain that allows them to interact with and regulate the Hsp70 family of molecular chaper- Current address for G. Wolber: Freie Universitat€ Berlin, Institute of Phar- macy, Department Pharmaceutical and Medicinal Chemistry, Berlin, ones (12, 13). BAG-1 can bind to the kinase domain of Germany. CRAF (14) or BRAF (15), and in vitro experiments showed Corresponding Author: Jakob Troppmair, Innsbruck Medical University, that BAG-1/CRAF interaction leads to the activation of Innrain 66, Innsbruck 6020, Austria. Phone: 43-512-504-27819; Fax: 43- CRAF independently of RAS (14). RAS-independent RAF 512-504-24624; E-mail: [email protected] activation, which results in the activation of ERK1/2, is also doi: 10.1158/1535-7163.MCT-13-0142 observed in cells overexpressing BAG-1 (16). ERK1/2 acti- 2013 American Association for Cancer Research. vation can be terminated by stress-induced upregulation

2400 Mol Cancer Ther; 12(11) November 2013

Thio-2 Inhibitor of BAG-1–Mediated Protein Interactions

of Hsp70 (16). Mechanistically, this may be explained by (42). Thioflavin S also reduces viability of wild-type (wt) the competition between CRAF and Hsp70 for a common but not BAG-1–deficient mouse embryonic fibroblasts binding site in helix 2 of the BAG domain. Genetic dis- (42). Until now, hit-to-lead development of drug-like ruption of BAG-1 in mice has no effect on the activation of inhibitors has been precluded by complexity of the Thio- ERK1/2 by mitogens (15). Nevertheless, subcellular local- flavin S mixture. Here, we report the isolation and puri- ization studies have suggested that BAG-1 forms a mito- fication of the compounds Thio-2, Thio-3, Thio-5, and chondrial survival signaling complex with Hsp70/AKT Thio-6 from Thioflavin S and their structural characteri- and BRAF, the absence of which may account for zation. We also demonstrate the ability of Thio-2 to block increased apoptosis in the liver and developing nervous signal propagation via the MAPK pathway, to impede the system of BAG-1–deficient mouse embryos (15). Evidence proliferation of RAF-transformed cells, and inhibit the for BAG-1 involvement in RAF-driven transformation has interaction of BAG-1 with Hsc70 and to a lesser degree been provided by the demonstration that BAG-1 hetero- RAF in HEK293 cells. zygosity in mice expressing a constitutively active form of CRAF in type II pneumocytes significantly reduces oncogene-induced lung adenoma growth (17). In this model, Materials and Methods RAF-downstream signaling was unaffected, whereas Reagents reduced BAG-1 expression specifically targeted tumor General chemicals were of molecular biology grade and cells to apoptosis (17). were purchased either from Sigma-Aldrich or Fisher Activating mutations in RAF kinases are almost exclu- Scientific, unless otherwise stated. Thioflavin S practical sively restricted to BRAF and are present in approximate- grade (CAS 1326-12-1) was purchased from Sigma- ly 10% of all human tumors with the highest incidence of Aldrich. Calculation of molarity was based on informa- approximately 42% (Catalogue of Somatic Mutations in tion provided by the National Cancer Institute Develop- Cancer Database, Wellcome Sanger Trust) detected in mental Therapeutics Program (NCI-DTP; Rockville, MD; melanoma (1, 5, 18). Most commonly BRAF mutant mel- http://dtp.cancer.gov). Analytical grade reagents (n-hex- anoma are V600E (74%–90%; ref. 19) and 16% to 29% are ane, diethyl ether, ethyl acetate, n-butanol, petroleum V600K mutations (20, 21). Because of the role of RAF ether, acetone, dichloromethane, and methanol) were kinases in promoting cancer cell growth, mainly through supplied by VWR. High-performance liquid chromato- enforcing cell-cycle progression and enhancing cell sur- graphy (HPLC)-grade acetonitrile and methanol were vival (1, 5, 18, 22, 23), the RAS–RAF–MEK module has obtained from Merck. Deionized water (18.2 MO cm) become a promising target for therapeutic intervention. was obtained from an Arium 611 UV system (Sartorius This led to the development of small molecular weight Stedim Biotech GmbH). The MEK-inhibitor U0126 was inhibitors of RAF and MAP–ERK kinase (MEK; refs. 5, 24), purchased from Promega Corporation and prepared as a with recent clinical studies reporting that highly specific 10 mmol/L stock solution in dimethyl sulfoxide (DMSO; BRAF inhibitors are effective in the treatment of metastatic Sigma-Aldrich). AZD6244 (selumetinib) was obtained melanoma (25–27). However, initial promise has been from Eubio and prepared as a 10 mmol/L stock in DMSO. hampered by the development of resistance (28–30), Staurosporine (100 mmol/L stock) was obtained from which is characterized by the reactivation of ERK1/2 Sigma-Aldrich. RAF inhibitors sorafenib (BAY43-9006) (31–33) and has been attributed to various mechanisms and PLX4032 were obtained from Axon Medchem BV including activating NRAS mutations (29), CRAF over- and prepared as 50 mmol/L and 10 mmol/L stocks in expression (34), compensatory upregulation of MAP2K DMSO, respectively. Thioflavin S (Sigma-Aldrich) and kinase COT (28), activating MEK1 mutations (35), and purified compounds (Thio-2, Thio-3, and Thio-5; for iso- amplification of mutant BRAF (36). lation and purification see below) were prepared as a 10 Disruption of specific protein–protein interactions (PPI) mmol/L stock solution in DMSO. All inhibitors were within signalosomes may provide an alternative approach, finally diluted in culture medium to reach working con- in addition to existing RAF protein conformation–specific centrations. Cell culture media and supplements were inhibitors, to interfere with aberrant signaling in cancer from PAA Laboratories, unless otherwise stated. The ECL cells. Indeed, it has been shown over the past few years that reagents (SuperSignal West Pico and Femto Chemilumi- PPI surfaces are druggable by small molecules (37–39), and nescent Substrates) were from Pierce Biotechnology. Glu- the first PPI inhibitors are now entering the clinic (39). tathione Sepharose 4B beads were from GE Healthcare Thioflavin S (NSC71948) is a complex mixture of dif- Bio-Sciences AB. Protein concentrations were routinely ferent components resulting from the methylation and assessed using Bio-Rad DC Protein Assay (Bio-Rad Lab- sulfonation of primulin base, a derivative of dehydrothio- oratories). Etoposide (Sigma-Aldrich) was prepared as a p-toluidine (40, 41). Thioflavin S interferes with the ability 20 mg/mL stock in DMSO. Formaldehyde and glutaral- of BAG-1 to interact with Hsc70 in vitro and in human dehyde (50% aqueous solution) were both obtained from breast cancer cell lines it reduces the interaction of BAG-1 Sigma-Aldrich. X-Gal was purchased from GIBCO, Life with Hsc70, Hsp70, and CRAF, without affecting inter- Technologies. Carboxyfluorescein diacetate succinimidyl action between BIM and MCL-1, and results in decreased ester (CFSE) was obtained from Molecular Probes, Life ERK1/2 phosphorylation, cell proliferation, and viability Technologies and stored at a 10 mmol/L stock in DMSO.

www.aacrjournals.org Mol Cancer Ther; 12(11) November 2013 2401

Enthammer et al.

Isolation and structural characterization of minutes back to ACN/H2O (20:80) and isocratic for 5 Thioflavin S-derived compounds Thio-2, -3, -5, minutes]. The detection was performed at 200, 254, 280, and -6 350, and 400 nm. Thioflavin S practical grade (CAS 1326-12-1; ca. 450 g) Fourier transform infrared spectroscopy (FTIR) analy- was first soxhlet extracted with 1 l diethyl ether and the sis. Infrared spectra were recorded on a Bruker (Bruker remaining powder extracted with water and diethyl ether to Optics) IFS 25 FTIR spectrometer in transmission mode yield 8.5 g crude extract, which was fractionated over silica (4,000–600 cm 1) using ZnSe disks of 2-mm thickness. gel (for further details see Supplementary Materials and Melting points were obtained on a Kofler hot-stage instru- Methods). The crystalline fractions were purified over an ment (uncorrected). SPE-C18 cartridge based on the color of the eluting bands; the eluate was divided into 10 and 12 subfractions, respec- Expression constructs tively. The subfractions eluting at CH3OH/H2O (90:10; 26.0 The pLNC-HA-MEK2-(S220D, S226D; constitutively and 16.6 mg, respectively) were combined and finally active) and pMCEF-myc-BRAF wt constructs were kindly purified over a Sephadex LH-20 column to yield 32.2 mg provided by Prof. Piero Crespo (Universidad de Cantab- of compound Thio-3. Fraction F4 (219.1 mg) was purified ria, Cantabria, Santander, Spain) and Prof. Richard Marais over an SPE-C18 cartridge. The subfraction eluting at (The Patterson Institute for Cancer Research, The Univer- dichloromethane/acetone 85:15 (48.7 mg), was finally puri- sity of Manchester, Manchester, UK), respectively. The fied by Sephadex LH-20 CC to yield 39.0 mg of compound HA-mBAG-1S construct was a kind gift of Prof. John Reed Thio-6. Fraction F5 (429.5 mg) was fractionated over an SPE- (Sanford-Burnham Medical Research Institute, La Jolla, C18 cartridge. The subfraction, which eluted at CH3OH/ CA). To generate pGEX-2TK-BAG-1S H3AB, BAG-1S H2O (85:15; 12.4 mg) was identified as Thio-2. Fraction F-6 H3AB was excised from pcDNA3-BAG-1S H3AB (43) by (534.5 mg) was fractionated over an SPE-C18 cartridge. The HindIII and XhoI double digestion and ligated into pGEX- subfraction, which eluted at CH3OH/H2O (80:20 and 85:15; 2TK in between the same flanking restriction sites. 70.8 mg), was also recognized as Thio-2. The powder-like crystals (318.1 mg), which precipitated from fraction F-8 Cell culture (367.9 mg), were combined with fraction F-10 (119.1 mg). MCF-7, MCF-10A, and HEK293 cells were obtained The combined fraction was purified over an SPE-C18 car- from LGC Standards. MCF-10A cells are derived from tridge. The subfraction eluted with dichloromethane/ace- adherent cells from a mammary epithelial cell line pro- tone 85:15 (v/v; 105.31 mg) was finally purified by Sepha- duced from long-term culture in serum-free medium with þ dex LH-20 CC to yield 35 mg of compound Thio-5. low Ca2 concentration (44). In contrast with MCF-7 Liquid chromatography/mass spectrometry analy- estrogen receptor–positive breast cancer cells (45), these sis. Electrospray–mass spectrometry (ESI–MS) spectra cells are nontumorigenic in immunosuppressed mice (46). were measured in ESI-positive mode on a Bruker (Bru- Mouse fibroblast cell lines NIH 3T3 (wt) and NIH 3T3 kerDaltonics) Esquire 3000 plus ion-trap mass spectrom- expressing constitutively active forms of BRAFV600E eter in positive mode (spray voltage þ4.5 kV; endplate (kindly provided by Prof. R. Marais; ref. 47) were grown offset 500 V; nebulizer gas 40 psi; drying gas flow in Dulbecco’s Modified Eagle Medium (DMEM). MCF-7 rate 9.00 L/min; dry temperature 350C; and m/z range cells were maintained in minimum essential medium 100–1,500) coupled to an HPLC Hewlett Packard HP (Sigma-Aldrich), HEK293 cells were kept in DMEM. 1100 instrument with autosampler, diode-array detector, UACC257 cells (provided by Dr. Roland Houben, Uni- and column thermostat at 40C using a Gemini C18 110A versity of Wurzburg,€ Wurzburg€ Germany) were kept in column (5 mm; 150 3.00 mm; Phenomenex) with guard RPMI, MDA-MB-231 and MDA-MB453 (LGC standards) column at a flow rate of 400 mL/min using an acetonitrile cells were maintained in DMEM. All media contained 10% gradient in water both containing 0.1% formic acid (same fetal calf serum (FCS), 2 mmol/L L-glutamine and 100 as for HPLC analysis). penicillin/streptomycin. For MCF-10A, cell culture HR-MS analysis. High-resolution ESI–MS spectra DMEM/F12 (1:1 v/v) medium (GIBCO, Life Technolo- were recorded on a micrOTOF-Q II mass spectrometer gies) was supplemented with 5% FCS, 20 ng/mL EGF, 0.5 Bruker (BrukerDaltonics) in positive mode (spray voltage mg/mL hydrocortisone, and 10 mg/mL ITS (insulin–trans- þ4.5 kV; endplate offset 500 V; nebulizer gas 10 psi; ferrin–sodium selenite). Parental (M229 and M238) and drying gas flow 5 L/min; dry temperature 180C; and m/z PLX4032-resistant (M229 R5 and M238 R1) melanoma cell range 100–1,000). lines (29) were provided by Dr. Antoni Ribas (University HPLC analysis. Analytical HPLC separations were of California, CA) and were maintained in complete cell performed on a HPLC Hewlett Packard HP-1050 system culture media or media supplemented with 1 mmol/L equipped with autosampler, diode-array detector, and PLX4032, respectively. All cell lines were cultured inhib- column thermostat at 40C using a Gemini C18 110A col- itor free for 96 hours before use. Cells were maintained at m umn (5 m; 150 3.00 mm; Phenomenex) with guard 37 C in a 5% CO2 humidified atmosphere. The cells were column at a flow rate of 400 mL/min using an acetonitrile passaged for less than 6 months in our laboratory after gradient in water containing 0.1% formic acid [ACN/H2O receipt or resuscitation. No authentication was done by (20:80) to (98:2) in 25 minutes, isocratic for 10 minutes, in 5 the authors.

2402 Mol Cancer Ther; 12(11) November 2013 Molecular Cancer Therapeutics

Thio-2 Inhibitor of BAG-1–Mediated Protein Interactions

Transfection procedures and HA-tag (12013819001; Roche Diagnostic). Densito- Cells were transfected using Lipofectamine 2000 (Life metric analyses were performed using Image J software Technologies-Invitrogen) according to the manufacturer’s (http://rsbweb.nih.gov/ij/). PPI was additionally test- protocol. ed using recombinant bacterially expressed and purified Hsc70, which had been labeleled with Fluorescein MTT assay isothiocyanate (FITC) and recombinant GST-BAG-1. MTT assay was carried out as previously described (48). Binding was determined spectrophotometrically using Absorbance was measured at 550 nm with a 620 nm a Varioskan Flash reader with SkanIt RE 2.4.3 Software reference filter using an Anthos 2010 Microplate Reader (Thermo Scientific). For further details see Supplemen- (Biochrom, Holliston, MA). tary Materials and Methods.

Determination of cell proliferation and viability Molecular modeling and docking experiments Cell proliferation was determined using a Neubauer Protein preparation was using the software program counting chamber or by performing crystal violet staining MOE (51). Docking studies were performed with GOLD (49). Cell viability was assessed by trypan-blue exclusion version 5.1 (GOLDScore, 100% search efficiency; ref. 52) assay of adherent and nonadherent cells. and without specific constraints. Protein Data Bank (PDB) (53) entry 1HX1 (54) was selected as protein template. Clonogenic assay Docking results were postprocessed using the software MCF-7 and MCF-10A cells were seeded (1 103 cells/ package LigandScout (55–57). For protein and binding site well) in triplicate and treated for 24 or 48 hours as visualization VMD (58) and LigandScout were used. described. Cells were kept in complete growth medium until colonies were visible to the naked eye. Colonies were Statistical analysis fixed and stained by adding 1.5 mL of a 0.5% crystal violet All data are presented as mean SD unless otherwise (Sigma-Aldrich) solution (in 50% methanol) for 45 min- stated. Statistical analysis was performed using Graph- utes and excess stain removed by immersing plates in Pad Prism 5 Software (GraphPad Software). One-way water. Plates were air-dried and colonies counted with a ANOVA and Tukey Multiple Comparison Test were used manual counter. for comparison of significant effects of one independent variable in multiple groups. To calculate statistical differences of multiple independent variables in two groups, Cellular signaling and protein–protein interaction two-way ANOVA with Bonferroni posttest was applied. Protein separation and immunoblot analysis was per- Significance values were designated as follows: , P < 0.05; formed as described previously (50). Antibodies raised , P < 0.01; , P< 0.001. against the following antigens were used: pERK (sc- 16982R) and ERK (sc-94; Santa Cruz Biotechnology); PARP (9542), caspase-7 (9492), pAKT (4058), AKT (4685), and Results 9E10 myc-tag (2276; Cell Signaling Technology); BRAF (sc- Isolation of the active compounds of Thioflavin S 5284, Santa Cruz Biotechnology) and GAPDH (glyceral- Because Thioflavin S is a mixture of two major dehyde-3-phosphate dehydrogenase; AM4300; Ambion). constituents [2-(4-(diethylamino)phenyl)-6-methylbenzo- For coimmunoprecipitation, HEK293 cells stably expres- [d]thiazole-7-sulfonic acid and 20-(4-(diethylamino)phe- sing wt HA-mBAG-1S were seeded in triplicate in a 6-well nyl)-6-methyl-2,60-bibenzo[d]thiazole-7-sulfonicacid;Rea- plate at a density of 6.5 105 cells per well. Samples were xys Registry Number 22509892; Fig. 1A], plus many more transiently transfected with equal amounts of pMCEF- compounds at lower concentrations, we used a simplified myc-BRAF. Thirty hours posttransfection cells were trea- bioguided isolation protocol to identify the compounds ted with Thio-2, Thioflavin S (50 and 100 mmol/L), or responsible for the biologic activity of Thioflavin S (37). DMSO for 16 hours and cells were lysed in HKEM buffer Thioflavin S was fractionated by means of liquid–liquid [50 mmol/L HEPES, pH 7.2, 5 mmol/L MgCl2,142 extraction with different solvents yielding five subfrac- mmol/L KCl, 2 mmol/L EGTA, and 0.2% (v/v) Nonidet tions. The highest growth inhibitory activity was found in P40] supplemented with 1:100 protease inhibitor. To pull the most lipophilic subfractions (MTT assay, data not down HA-mBAG-1S, 350–500 mg total protein was incu- shown). Furthermore, extraction yielded four pure com- bated with 30 mL of anti-HA–affinity matrix (Roche Diag- pounds: Thio-2, -3, -5, and -6 (Fig. 1A). Additional sub- nostic) overnight on an orbital shaker at 4 C. Immune stantive information is provided in Supplementary Tables complexes were washed five times with HKEM buffer, S1–S4 and Supplementary Figs. S1–S4. beads were resuspended in 50 mL2 Laemmli buffer The stability of Thio-2, -3, -5, and -6 under tissue culture and heated for 5 minutes at 95 C. Immunoblot analysis conditions was determined by LC-ESI–MS analysis and was performed as described previously (50). Antibodies detected no decline of compounds Thio-2, -5, and -3 raised against the following antigens were used: BAG-1 (purity >98%) after 72 hours. However, as degradation (sc-8348) and Hsc70 (sc-7298; Santa Cruz Biotechnolo- of Thio-6 was seen (data not shown), this compound was gy); 9E10 myc-tag (2276; Cell Signaling Technology); not investigated further.

www.aacrjournals.org Mol Cancer Ther; 12(11) November 2013 2403

Enthammer et al.

Figure 1. Properties of Thioflavin S and purified compounds and their biologic characterization. A, known major constituents of Thioflavin S (1, 2) and Thioflavin S-derived compounds: Thio-2 (N-ethyl-4-(6-methylbenzo[d]thiazol-2-yl)aniline), Thio-3 (N,N-diethyl-4-(6-methylbenzo[d]thiazol-2-yl)aniline), Thio- 5(N-ethyl-4-(6-methyl-[2,60-bibenzo[d]thiazol]-20-yl)aniline), and Thio-6 (N,N-diethyl-4-(6-methyl-[2,60-bibenzo[d]thiazol]-20-yl)aniline). B, effect of Thioflavin S and purified compounds on cell growth. (Continued on the following page.)

2404 Mol Cancer Ther; 12(11) November 2013 Molecular Cancer Therapeutics

Thio-2 Inhibitor of BAG-1–Mediated Protein Interactions

Cell growth inhibition and apoptosis induction by MCF-10A cells (Fig. 2 and Supplementary Fig. S6). Serum Thioflavin S-derived compounds stimulation resulted in a pronounced phosphorylation of To test the biologic activity of Thioflavin S and the ERK1/2, which was effectively blocked by MEK inhibi- pure compounds Thio-2, Thio-3, and Thio-5, MCF-7 tors U0126 and AZD6244 used as a positive control in both cells were treated with 0 to 100 mmol/L of these com- cell lines (Fig. 2A). Thio-2 in contrast only blocked ERK1/2 pounds for 72 hours. Treatment with Thioflavin S activation in MCF-7, whereas in MCF-10A cells even caused a 14% growth inhibition at 100 mmol/L as mon- increased phosphorylation is observed. However, Thio- itored by the MTT assay (Fig. 1B). Of note, 100 mmol/L 2 had no significant effect on the activation of AKT (a Thio-3 and -5 caused 26% and 38% cell growth inhibi- prosurvival kinase downstream of RAS but outside the tion, respectively. MCF-7 cells maintained in the RAF–MEK–ERK pathway) in MCF-7 cells, whereas AKT presence of Thio-2 proliferated significantly slower phosphorylation was decreased in MCF-10A cells (Fig. compared with control cells (P < 0.001) or Thioflavin 2A). Specificity for RAF-dependent signaling in trans- S–treated cells (P < 0.001). Cell proliferation was inhib- formed cells was also supported by experiments using ited in a dose-dependent manner with a mean IC50 of a combination of Thio-2 and an inhibitor of RAF. As 29.6 7 mmol/L for Thio-2. The growth inhibitory effect shown in Fig. 2B, application of 5 mmol/L sorafenib, an of Thio-2 was not limited to MCF-7 cells but was inhibitor of CRAF and BRAF proteins, caused a modest observed in two other breast cancer cell lines (MDA- reduction (32%) in ERK1/2 phosphorylation, whereas the MB-231 and MDA-MB-453) and the melanoma cell line combination of 5 mmol/L sorafenib with 25 mmol/L Thio- UACC257 (Supplementary Fig. S5). 2 significantly reduced ERK1/2 phosphorylation (P < The number of viable cells was also counted 72 hours 0.01) compared with sorafenib alone (Fig. 2C). These following treatment with 25 mmol/L of each compound findings suggest that Thio-2 may act at the level of RAF using U0126 (an inhibitor of the RAF effector kinase or MEK. To further test this hypothesis, HEK293 cells MEK) as a positive control (Fig. 1C). Significantly overexpressing a constitutively active mutant form of reduced cell numbers were observed after 72 hours in MEK (MEK2S220D/S226D; ref. 59) were treated with the the presence of Thioflavin S, Thio-2, -3, and -5, whereas MEK-specific inhibitor U0126 or Thio-2. ERK phosphor- the effect of U0126 became apparent at 48 hours ylation was inhibited by U0126 but was unaffected by (Fig. 1C). Compared with other Thio fractions, Thio-2 Thio-2 (Fig. 2D). Finally, we also directly tested for the exhibited the most pronounced inhibition by 72 hours effect of sorafenib or Thio-2 on the phosphorylation of (Fig. 1C). MEK following serum stimulation of MCF-7 cells. As Cell death induced by Thio-2 treatment was analyzed shown in Fig. 2E, Thio-2 efficiently prevented MEK phos- by monitoring proteolytic processing of caspase-7 and phorylation. All these findings are consistent with an PARP, which was highest with Thio-2 concentrations inhibitory function of Thio-2 in the RAF–MEK–ERK path- 25 mmol/L (Fig. 1D). In contrast, no PARP cleavage was way, most likely at the level of RAF. observed with Thioflavin S (Fig. 1D), Thio-3, and Thio-5 at concentrations up to 100 mmol/L (data not shown). A Thio-2 selectively targets transformed cells clonogenic survival assay was also performed using 100 Thio-2 12 mmol/L significantly inhibited growth of mmol/L of Thio-2 or Thioflavin S (Fig. 1E). Staurosporine transformed MCF-7 cells in a dose-dependent manner was included as a positive control. Thio-2 potently compared with untransformed MCF-10A cells (Fig. 3A). reduced colony yield compared with Thioflavin S or This difference in the response to Thio-2 was also obvious DMSO control. On the basis of this initial characterization, when activation of apoptotic signaling was tested by Thio-2 was selected for all subsequent analyses. immunoblotting for the processing of caspase-7 and PARP (Fig. 3B or when a clonogenic survival assay was Thio-2 targets the RAF–MEK–ERK signaling performed (Fig. 3C and D). Quantification of trypan-blue pathway stained cells (nonviable) confirmed the pronounced sen- To gain insights into possible regulation of RAF by sitivity of MCF-7 to Thio-2 treatment compared with Thio-2, we studied its effects on the activation of MCF-10A cells (Supplementary Fig. S7). As tumor cells ERK1/2 in MCF-7 cells compared with untransformed may carry multiple genetic alterations, NIH 3T3 murine

(Continued.) MCF-7 cells were incubated for 72 hours with the compounds indicated or DMSO and analyzed using the MTT assay. Thio-6 was unstable under the chosen conditions and was not further investigated. Data shown are mean SD and derived from at least three independent experiments, each carried out in triplicate. , P < 0.001 refers to significantly different from DMSO control and Thioflavin S. C, cell growth was determined every 24 hours over three consecutive days. Briefly, MCF-7 cells were seeded at a density of 2 105 cells per well and treated with the indicated substances at 25 mmol/L. The U0126 MEK inhibitor served as a positive control. Data shown (mean SD) are derived from four independent experiments. , P < 0.05; , P < 0.01; , P < 0.001, refers to significantly different from DMSO control. D, MCF-7 cells were analyzed for apoptotic cell death after 48 hours of treatment based on the processing of PARP or caspase-7. GAPDH served as a loading control. A single representative experiment is shown (n ¼ 3). E, effect of Thio-2 and Thioflavin S on clonogenic cell survival. MCF-7 cells were seeded in triplicate (1 103 cells/well) and treated for 24 or 48 hours with the compounds indicated. Cells were kept in complete growth medium until cell colonies became visible to the naked eye. Statistical significance was determined using one- way ANOVA and is expressed compared with DMSO control. Treatments were performed in triplicate and data shown are derived from two different experiments; , P < 0.05; , P < 0.01. www.aacrjournals.org Mol Cancer Ther; 12(11) November 2013 2405

Enthammer et al.

Figure 2. Thio-2 targets the MAPK signaling pathway. A, densitometric analyses of ERK1/2 and AKT phosphorylation in MCF-7 and MCF-10A cells. Experiments were carried out in reduced serum medium (0.5% FCS) and lysates were prepared after 10 minutes of serum stimulation (10% FCS). Cells were kept in reduced serum media for 6 hours and treated with the substances indicated for 1 hour. Activation of relevant pathways was analyzed using phosphospeciﬁc antibodies. (Continued on the following page.)

2406 Mol Cancer Ther; 12(11) November 2013 Molecular Cancer Therapeutics

Thio-2 Inhibitor of BAG-1–Mediated Protein Interactions

fibroblast cell line–expressing wt or the oncogenic V600E 0.8, to accurately measure the effect of Thio-2. Glutathione mutant human BRAF proteins were examined. Cell S-transferase (GST)-tagged BAG-1S, but neither the neg- growth was inhibited by Thio-2 in a dose-dependent ative control GST-BAG-1S H3AB nor GST alone interacted manner in BRAFV600E but not wt cells as shown by try- with FITC-labeled Hsc70 (typical degree of labeling was pan-blue exclusion and crystal violet cell–staining assays around 0.5 mol of FITC/mole of Hsc70). GST-tagged (Fig. 3E and F, respectively). Immunoblotting showed no proteins did not interact with the unrelated protein increase in the processing of PARP or caspase-3 cells in Annexin V–FITC (Fig. 4C). Thioflavin S dose-dependently BRAFV600E-expressing fibroblasts following treatment inhibited interaction between GST-BAG-1S and Hsc70- with 25 to 100 mmol/L Thio-2 (Supplementary Fig. FITC with DMSO (vehicle) alone exhibiting no effect S8A), indicating the absence of apoptotic cell death under on binding; as expected there was negligible binding these conditions. Also, a b-gal activity assay has been between GST-BAG-1S H3AB and Hsc70-FITC at concen- carried out but yielded no evidence that Thio-2–treated trations of Thioflavin S up to 100 mmol/L (Fig. 4D). Thio-2 cells show increased senescence (data not shown). Finally, dose-dependently targeted GST-BAG-1S/Hsc70-FITC CFSE staining hinted decreased proliferation in U0126- interaction, causing a mean SEM inhibition of binding and Thio-2–treated cells as obvious from the comparison of 28.45 7.15 at a concentration of 75 mmol/L (Fig. 4E). of histogram plots of DMSO- (green line) and Thio-2/ To corroborate our in vitro data, we further analyzed U0126–treated (red lines) cells (Supplementary Fig. S8B). PPIs in cells. In HEK293 cells stably expressing HA- NIH 3T3 cells are characterized by low basal phosphor- mBAG-1S, Thioflavin S, and Thio-2 significantly decreased ylation of ERK1/2, which is, however, pronounced in BAG-1 binding to Hsc70 over the dose range applied (Fig. BRAF-transformed cells due to constitutive RAF signaling 4F and Supplementary Fig. S10). This effect extended to the (Fig. 3G). Phosphorylation of ERK1/2 was efficiently BAG-1/myc-BRAF interaction, which was also decreased reduced in BRAFV600E-expressing cells by Thio-2 but not upon incubation with Thioflavin S or Thio-2. in wt fibroblasts (Fig. 3G and H). When combined together, Taken together our results confirm the ability of Thio-2 the MEK-inhibitor U0126 and Thio-2 had an additive effect to disrupt BAG-1–mediated protein interactions and in suppressing ERK1/2 phosphorylation in BRAFV600E illustrate a possible mode of action of this Thioflavin cells (Fig. 3I). To exclude the possibility that the effect of S derivative. Thio-2 treatment is caused through BRAF destabilization, BRAF protein expression was analyzed by immunoblot- Melanoma cells resistant to PLX4032 respond to ting. No decrease in endogenous or mutant BRAF levels Thio-2 treatment was observed in MCF-7 or NIH 3T3 cells after treatment The potential of Thio-2 treatment to overcome melano- with Thio-2 for up to 48 hours (Supplementary Fig. S9C) or ma resistance to PLX4032, a drug that targets oncogenic 72 hours (Supplementary Fig. S9A and S9B), respectively. mutant BRAF with high specificity (29), was examined using drug-responsive (M229 and M238) or -resistant Thio-2 targets BAG-1–mediated protein interactions (M229 R5 and M238 R1) cell lines. Thio-2–inhibited cell We investigated whether purified Thio-2 could inhibit growth dose-dependent and this was statistically signif- the interaction of the cochaperone BAG-1 with Hsc70 and icant compared with DMSO control at concentrations 50 BRAF. Thioflavin S (1 mmol/L) but not its structurally mmol/L in all melanoma cell lines tested (Fig. 5A). In related analogs Thioflavin T (1 mmol/L) or BTA-1 (1 PLX4032-resistant M229 R5 cells, growth was reduced mmol/L), inhibited interaction between BAG-1S and more effectively compared with controls. Moreover, the Hsc70; none of the compounds affected protein integrity degree of inhibition achieved with Thio-2 in PLX4032- (Fig. 4A). Recombinant BAG-1S H3AB mutant protein, resistant M229 R5 cells was comparable with that of which is defective in binding Hsc70, was used to confirm PLX4032 in M229 cells (Fig. 5B). We also analyzed the specificity of interaction between BAG-1S and Hsc70. effects of Thio-2 on ERK1/2 activation (Fig. 5C). To elim- Thio-2–inhibited GST-BAG-1S/Hsc70 interaction relative inate stimulatory effects of serum growth factors cells to Thioflavin T control, but its effect was not as pro- were maintained at 0.05% serum. All cell lines exhibited nounced as that of Thioflavin S (Fig. 4B). robust ERK1/2 phosphorylation, which was efficiently As the concentration of compounds required to inhibit decreased by the treatment with PLX4032 in both, resis- in vitro BAG-1S/Hsc70 interaction was quite high, an tant and responsive cells, as reported previously (29) and ELISA assay was developed, with a Z score range 0.5 to also to a lesser degree by Thio-2.

(Continued.) Statistical signiﬁcance was calculated by applying one-way and two-way ANOVA, respectively (mean SD; , P < 0.05; , P < 0.01). B, effect of the combined application of sorafenib and Thio-2 (1 hour) on the phosphorylation of ERK1/2 in MCF-7 cells. Experiments were carried out under low-serum conditions and lysates were prepared after 10 minutes of serum stimulation. C, densitometric analysis of data shown in B. Phosphorylation levels of single and joint treatments are expressed relative to DMSO control (100%, mean SD). , P < 0.01, refers to signiﬁcantly different from DMSO control. D, activated MEK overcomes the block of ERK1/2 phosphorylation by Thio-2. Signaling analysis was conducted 48 hours after transfection as described in A. Expression of transiently transfected HA-MEK2S220D/S226D in MCF-7 cells was monitored using HA-tag antibody. All blots shown here are representative of three independent experiments. E, immunoblot analysis of MEK1/2 phosphorylation. MCF-7 cells were kept in reduced serum media for6 hours and treated with the substances indicated for 1 hour. Lysates were obtained 10 minutes after of serum stimulation.

www.aacrjournals.org Mol Cancer Ther; 12(11) November 2013 2407

Enthammer et al.

Figure 3. Effect of Thio-2 on growth, survival, and signaling of untransformed and transformed cells. A, a total of 1 104 MCF-7 and MCF-10A cells were seeded and incubated for 72 hours with Thio-2 or DMSO and analyzed by MTT assay. Data show the mean SD from at least two independent experiments, each performed in triplicate. Statistical significance was determined by two-way ANOVA. , P < 0.05; , P< 0.001. B, proapoptotic effects of Thio-2. Cells were analyzed for apoptotic cell death signaling after a 48 hours treatment by monitoring the processing of PARP and caspase-7 after treatment with the indicated compounds. GAPDH served as a loading control. A single representative experiment is shown (n ¼ 3). C and D, effect of Thio-2 and Thioflavin S on clonogenic cell survival. MCF-7 and MCF-10A cells were seeded (1 103 cells/well) in triplicate and treated for 24 (C) or 48 (D) hours, with the compounds indicated. Cells were kept in complete growth medium until cell colonies were visible to the naked eye and stained as described in Materials and Methods. The mean number of colonies is depicted as a percentage of DMSO control (100%). Data are represented as mean SD (n ¼ 2). Statistical significance was determined by two-way ANOVA. (Continued on the following page.)

2408 Mol Cancer Ther; 12(11) November 2013 Molecular Cancer Therapeutics

Thio-2 Inhibitor of BAG-1–Mediated Protein Interactions

Thio-2 presumably binds to BAG-1 and can result in growth arrest (16). It is therefore con- To further strengthen our hypothesis that primarily the ceivable that protein competition for binding to BAG-1 binding of Thio-2 to BAG-1 is the reason for the effects may occur alongside the inhibitory effect of Thio-2 or described, we performed molecular modeling experi- Thioflavin S thereby reducing binding of Hsc70 or RAF ments starting from the crystal structure of the complex to BAG-1 in cells. of BAG-1 and Hsc70. On the basis of structural and Our docking experiments show that Thio-2 potentially mutational studies published by Sondermann and collea- binds to the BAG domain of BAG-1 at the binding inter- gues (54), a putative binding site for small-molecule face of Hsc70, thereby targeting PPIs (Fig. 6). Amino acids inhibitors was predicted in the interface of the BAG-1/ Asp222 (helix 2) and Arg237 (helix 3) within the BAG Hsc70 complex (Fig. 6A). Docking revealed plausible domain are critical for binding to the ATPase domain of binding poses for Thio-2, showing supramolecular inter- Hsc70 (54) and may serve as contact sites for Thio-2 actions with the C-terminal part of BAG-1. In particular, according to our model. The requirement of helices 1 and p-stacking of the aniline and benzothiazol moieties with 2 of the BAG domain for interaction with RAF has also Arg234 and Arg237 (helix 3 of the BAG domain), hydro- been shown (16), although the amino acids that are critical gen bonding of the aniline nitrogen with Asp222 (helix 2 of for binding at the protein interface have not been identi- the BAG domain), and a hydrophobic contact of the ethyl fied as yet. group with Ile225 could be observed (Fig. 6B). Notably, Thio-2 showed a pronounced inhibition of RAF signal- Asp222 and Arg237, which are evolutionarily highly ing, as is evident from the decreased phosphorylation of conserved, are essential for forming the interaction sur- ERK1/2 in cells stimulated with serum or expressing an face with the Hsc70 ATPase domain (54). oncogenic RAF mutant (Figs. 2, 3 and 5). We failed to detect any such effect of Thio-2 on AKT, which is also activated by mitogen stimulation and lies downstream of Discussion RAS. Thio-2 cooperated with another inhibitors of the We report the isolation and characterization of the MAPK pathway, such as the RAF-inhibitor sorafenib in Thioflavin S–derived compound Thio-2 (Fig. 1 and Sup- suppressing growth factor–induced ERK1/2 phosphory- plementary Data), which is mainly responsible for the lation. The inhibitory effect of Thio-2 on ERK1/2 activa- biologic activities and biochemical properties previously tion was overcome by the expression of a constitutively reported for Thioflavin S (42). In an ELISA-based assay active form of MEK (59), further supporting that Thio-2 using recombinant BAG-1 and Hsc70 proteins, Thio-2 was acts upstream of MEK, possibly at the level of RAF. able to decrease binding of Hsc70 to immobilized BAG-1 With regard to RAF signaling, two possible functions in a dose-dependent manner, although less efficiently have been proposed for BAG-1: activation of RAF in an than the parental compound Thioflavin S (Fig. 4 and RAS-independent fashion (14, 16) and mitochondrial relo- Supplementary Data). In contrast, comparable efficacies calization of RAF (15). The mechanistic details underlying for Thioflavin S and Thio-2 were observed when protein these effects remain elusive. In the general scheme of RAF binding was studied by coimmunoprecipitation in activation, no essential role has been demonstrated for HEK293 cells. These cell-based experiments demonstrat- BAG-1 by biochemical or genetic studies. Our data show ed that Thio-2 or Thioflavin S also targeted BAG-1/RAF that Thio-2 inhibits signaling downstream of RAF acti- interaction (Fig. 4 and Supplementary Data). The reasons vated either by serum growth factors or as a result of for the higher potency of the PPI inhibitors in cells are activating mutation, suggesting that RAF/BAG-1 inter- unclear. While in the in vitro experiments with recombi- action is not primarily required for RAF activation. How nant protein test substances were added to immobilized Thio-2 prevents signaling from active RAF remains to be BAG-1 prior to the incubation with Hsc70 (in the absence established. Analysis of BAG-1–deficient mice has dem- of RAF), in cells, complexes of BAG-1 with Hsc70 or RAF onstrated the requirement of a BAG-1–induced mitochon- (and presumably other partners) preexist and will also be drial RAF translocation for cell survival (15), suggesting formed de novo during incubation with the compounds. that localization of RAF is critical for determining signal- Competition between RAF and Hsp70 for binding to the ing outcome. The analysis of signaling through RAF/ BAG domain has been demonstrated (16) and other pro- MEK/ERK led to the identification of several proteins, teins may also bind to this region of BAG-1. Upregulation which are not members of the core-signaling machine, but of Hsp70 during cellular stress displaces RAF from BAG-1 are required for compartmentalized signaling. In the case

(Continued.) E, NIH 3T3 wt and BRAFV600E fibroblasts were seeded at a density of 2 105 cells and maintained in medium supplemented with Thio-2 at the indicated concentrations. After 72 hours incubation, floating as well as adherent cells were counted using the trypan-blue exclusion assay. Total numbers of viable cells are indicated as mean SD and were derived from four independent experiments. F, a total of 2 105 NIH 3T3 wt and BRAFV600E cells were seeded and incubated for 72 hours with Thio-2 at various concentrations. Adherent cells were washed twice with PBS and stained with crystal violet. Values (mean SD) are depicted in percentage of DMSO control (100%) and were obtained from five independent experiments. G, effect of Thio-2 on basal ERK1/2 phosphorylation in NIH 3T3 wt and NIH BRAFV600E cells. Cells were incubated for 1 hour with the compounds indicated. GAPDH served as loading control. H, densitometric analysis of data shown in G. Phosphorylation levels are expressed relative to DMSO control (100%, mean SD). Statistical significance was determined by two-way ANOVA. I, NIH 3T3 BRAFV600E cells were kept in starvation medium and treated with U0126 and Thio-2 as indicated for 1 hour.

www.aacrjournals.org Mol Cancer Ther; 12(11) November 2013 2409

Enthammer et al.

Figure 4. Thio-2 dose-dependent targets BAG-1S/Hsc70 interaction. A, immunoblot analysis of in vitro pull-down reactions shows Hsc70 protein bound to GST, GST-BAG-1S H3AB, or GST-BAG-1S in the presence or absence of Thioﬂavin S (1 mmol/L), Thioﬂavin T (1 mmol/L), or BTA-1 (1 mmol/L). The amount of GST-tagged proteins used in the pull-down is shown in the input. Data are representative of three independent experiments. (Continued on the following page.)

2410 Mol Cancer Ther; 12(11) November 2013 Molecular Cancer Therapeutics

Thio-2 Inhibitor of BAG-1–Mediated Protein Interactions

Figure 5. Thio-2 inhibits proliferation of PLX4032-resistant cells. A and B, PLX4032 responsive (M229 and M238) or resistant (M229 R5 and M238 R1) cells were treated with Thio-2 at the concentrations indicated and analyzed using the MTT assay after 72 hours. Data shown are mean SD and derived from at least three independent experiments, each carried out in triplicate. Signiﬁcance values were calculated for the differences in response between Thio-2- and DMSO-treated cells (A) or for the difference between PLX4032-responsive and -resistant cells (B). C, immunoblot analysis of ERK1/2 phosphorylationin melanoma cell lines. Cells were kept in serum-reduced (0.5% FCS) media for 6 hours and treated with the substances indicated for 5 hours. Activation of the MAPK pathway was analyzed using phosphospeciﬁc antibodies. GAPDH served as a loading control. of RAS/RAF transformation signaling from sites other proliferation, survival, and signaling of transformed than than the cell membrane has been suggested (60). It untransformed cells. The frequently increased BAG-1 remains to be shown whether BAG-1 is also involved in levels observed in tumors (61) may therefore be required localizing RAF to any of these sites. Our data also dem- for other signaling events in transformed cells in addition onstrate that Thio-2 has a more pronounced effect on to BAG-1–mediated antiapoptotic functions.

(Continued.) B, immunoblot analysis of in vitro pull-down assay shows the effect of Thioflavin S (1 mmol/L) or Thio-2 (1 mmol/L) on GST-BAG-1S/Hsc70 binding relative to Thioflavin T (1 mmol/L), which was used as negative control. Immunoblot analysis is representative of three independent experiments repeated in duplicate and densitometricanalysisshowsthemean SEM from these. C, in vitro ELISA assay shows binding between GST, GST-BAG-1S H3AB, and GST-BAG-1S with Hsc70-FITC or the unrelated protein AnnexinV–FITC. Data are representative of duplicate determinations from two independent experiments. D, dose-dependent effect of Thioflavin S on the binding of GST-BAG-1S H3AB or GST-BAG-1S to Hsc70-FITC. DMSO (5% v/v) was used as vehicle control to exclude any nonspecific effects exerted by the diluent. Data are representative of duplicate determinations from three independent experiments. E, ELISA assay shows the effect of increasing concentrations of Thioflavin S, Thio-2, or Thioflavin T on the interaction of GST-BAG-1S with Hsc70-FITC. Data shown are normalized to the GST-BAG-1S H3AB/Hsc70-FITC background interaction and are representative of duplicate determinations from at least three independent experiments. F, HEK293 cells stably expressing HA-mBAG-1S were seeded in triplicate, transiently transfected with pMCEF-myc-BRAF using lipofection, and subjected to Thio-2, Thioflavin S (50 and 100 mmol/L, respectively) or DMSO for 16 hours. HA-mBAG-1S was captured using HA-affinity matrix and pull-down was analyzed using antibodies directed against BAG-1, Hsc70, and myc-tag. Immunoblot analysis shown is representative of three independent experiments.

www.aacrjournals.org Mol Cancer Ther; 12(11) November 2013 2411

Enthammer et al.

Arg234 Figure 6. Docking experiments lle225 elucidate possible binding of Thio- 2 to BAG-1. A, indicated is a Arg237 putative binding pocket within a-helices 2 and 3 of the BAG Lys238 domain. Shown in red are amino acids comprising the predicted binding site. Thio-2 is presented in ball and stick representations and color coded according to chemical elements. a-helices of the BAG Asp222 domain are depicted in blue, the isosurface in gray. Thio-2 potentially binds to the binding interface of Hsc70. B, surmised interaction pattern of Thio-2 with Lys238 Arg237 lle225 BAG-1 in 3D (top) and 2D (bottom). Purple discs represent p-stacking, S yellow spheres hydrophobic contacts and the green arrow a NH hydrogen bond as calculated by N LigandScout.

Arg234 Asp222

Thio-2 was also tested in a panel of melanoma cells resistant to treatment with PLX4032, an inhibitor of harboring the common BRAFV600E mutation, which are mutant BRAF. We demonstrate that Thio-2 inhibits sig- responsive or resistant to PLX4032 a mutant BRAF-spe- naling at the level of RAF. The mode of action of Thio-2 cific inhibitor (Fig. 5). Thio-2 proved effective at inhibiting may depend on its ability to interfere with the binding of growth of both PLX4032-responsive and -resistant cells. In RAF and/or Hsc70 to BAG-1 and further work will be particular, M229 PLX4032–resistant cells (M229 R5) were required to delineate this. even more susceptible than PLX4032-responsive cells to Thio-2–induced growth inhibition. M229 R5 cells acquire Disclosure of Potential Conflicts of Interest resistance to PLX4032 by upregulating the platelet– No potential conflicts of interest were disclosed. derived growth factor receptor b (PDGFRb; ref. 29) allow- ing for the activation of alternative prosurvival pathways. Authors' Contributions Conception and design: G. Wolber, G. Packham, R.I. Cutress, H. Stuppner, M229 R5 and M238 R1 cells also remained highly resistant J. Troppmair to MEK inhibition, and only stable knockdown of Development of methodology: M. Enthammer, E.S. Papadakis, PDGFRb caused cell-cycle arrest and apoptosis (29). The M. Deutsch, M.I. Ashraf, G. Wolber, G. Packham, R.I. Cutress Acquisition of data (provided animals, acquired and managed patients, effect of Thio-2 in these cells, which are unresponsive to provided facilities, etc.): M. Enthammer, E.S. Papadakis, K. Koziel, further targeting of the RAF-MEK-ERK cascade (29), is S. Khalid, R.I. Cutress surprising. One possible explanation is based on the Analysis and interpretation of data (e.g., statistical analysis, biostatis- tics, computational analysis): M. Enthammer, E.S. Papadakis, M.I. Ashraf, observation that BAG-1 enhances PDGFR-mediated pro- G. Wolber, G. Packham, R.I. Cutress, J. Troppmair tection from apoptosis through direct binding to the Writing, review, and/or revision of the manuscript: M. Enthammer, E.S. Papadakis, M.S. Gachet, S. Schwaiger, K. Koziel, M.I. Ashraf, G. Wolber, receptor (62). The region involved on the side of BAG-1 G. Packham, R.I. Cutress, H. Stuppner, J. Troppmair again involves helices 2 and 3 of the BAG-domain, which Administrative, technical, or material support (i.e., reporting or orga- are also necessary for Hsc70 binding. Thio-2–mediated nizing data, constructing databases): M. Enthammer, E.S. Papadakis, M.S. Gachet, M. Deutsch, H. Stuppner, J. Troppmair inhibition of M229 R5 cell growth could be the result of the Study supervision: G. Wolber, R.I. Cutress, H. Stuppner, J. Troppmair disruption of BAG-1/PDGFR binding, pointing to a novel mechanism for overcoming BRAF-inhibitor resistance. Acknowledgments The authors thank Dr A. Ribas for the kind gift of PLX4032-respon- Conclusions sive and -resistant cell lines and Profs.R.Marais,P.Crespo,andJ.Reed for providing expression constructs. The authors also thank Dr. P. We have identified and extensively characterized Thio- Duriez in the Cancer Research UK Protein Core Facility, Cancer 2, which impedes the growth of cell lines that had become Sciences Unit, University of Southampton (Southampton, UK), for

2412 Mol Cancer Ther; 12(11) November 2013 Molecular Cancer Therapeutics

Thio-2 Inhibitor of BAG-1–Mediated Protein Interactions

protein expression, puriﬁcation, and labeling. The authors thank Dr. L. nologies by the Federal Ministry for Transport, Innovation, and Tech- Forse for critical reading of the article and all members of the J. nology (BMVIT) and the Federal Ministry of Economy, Family, and Troppmair Laboratory for helpful discussions. The authors also thank Youth (BMWFJ) as well as the federal states Tyrol and Styria. The the valuable assistance of Ruth Baldauf in the preparation of the Austrian Research Promotion Agency (FFG) manages the competence article. center program COMET. J. Troppmair was also supported by the FWF, Austrian Science Foundation, project MCBO ZFW011010-08. M. Enthammer is supported by a project from the Austrian Cancer Grant Support Society/Tyrol. The work was supported by a Cancer Research UK grant (South- The costs of publication of this article were defrayed in part by the ampton Cancer Research Centre Core Award grant, ref: C34999/ payment of page charges. This article must therefore be hereby marked A11344; R.I Cutress). The work in the laboratory of R.I. Cutress was advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate also supported by a Research Support grant from the Royal College of this fact. Surgeons of Edinburgh and by a project grant from Breast Cancer Campaign (ref 2011NovPR39). J. Troppmair obtained ﬁnancial support from Oncotyrol, Project 1.5. The competence center Oncotyrol is funded Received March 1, 2013; revised August 26, 2013; accepted September 9, within the scope of COMET—Competence Centers for Excellent Tech- 2013; published OnlineFirst September 18, 2013.

References 1. Zebisch A, Troppmair J. Back to the roots: the remarkable RAF 21. Long GV, Menzies AM, Nagrial AM, Haydu LE, Hamilton AL, Mann GJ, oncogene story. Cell Mol Life Sci 2006;63:1314–30. et al. Prognostic and clinicopathologic associations of oncogenic 2. Osborne JK, Zaganjor E, Cobb MH. Signal control through Raf: in BRAF in metastatic melanoma. J Clin Oncol 2011;29:1239–46. sickness and in health. Cell Res 2012;22:14–22. 22. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, et al. 3. Kolch W. Coordinating ERK/MAPK signalling through scaffolds and Mutations of the BRAF gene in human cancer. Nature 2002;417: inhibitors. Nat Rev Mol Cell Biol 2005;6:827–37. 949–54. 4. Matallanas D, Birtwistle M, Romano D, Zebisch A, Rauch J, von 23. Zebisch A, Staber PB, Delavar A, Bodner C, Hiden K, Fischereder K, Kriegsheim A, et al. Raf family kinases: old dogs have learned new et al. Two transforming C-RAF germ-line mutations identified in tricks. Genes Cancer 2011;2:232–60. patients with therapy-related acute myeloid leukemia. Cancer Res 5. Zebisch A, Czernilofsky AP, Keri G, Smigelskaite J, Sill H, Troppmair J. 2006;66:3401–8. Signaling through RAS-RAF-MEK-ERK: from basics to bedside. Curr 24. Chappell WH, Steelman LS, Long JM, Kempf RC, Abrams SL, Franklin Med Chem 2007;14:601–23. RA, et al. Ras/Raf/MEK/ERK and PI3K/PTEN/Akt/mTOR inhibitors: 6. Udell CM, Rajakulendran T, Sicheri F, Therrien M. Mechanistic prin- rationale and importance to inhibiting these pathways in human health. ciples of RAF kinase signaling. Cell Mol Life Sci 2011;68:553–65. Oncotarget 2011;2:135–64. 7. Roy F, Laberge G, Douziech M, Ferland-McCollough D, Therrien M. 25. Bollag G, Hirth P, Tsai J, Zhang J, Ibrahim PN, Cho H, et al. Clinical KSR is a scaffold required for activation of the ERK/MAPK module. efficacy of a RAF inhibitor needs broad target blockade in BRAF- Genes Dev 2002;16:427–38. mutant melanoma. Nature 2010;467:596–9. 8. Dumaz N, Marais R. Protein kinase A blocks Raf-1 activity by stimu- 26. Flaherty KT, Puzanov I, Kim KB, Ribas A, McArthur GA, Sosman JA, lating 14-3-3 binding and blocking Raf-1 interaction with Ras. J Biol et al. Inhibition of mutated, activated BRAF in metastatic melanoma. Chem 2003;278:29819–23. N Engl J Med 2010;363:809–19. 9. Dai P, Xiong WC, Mei L. Erbin inhibits RAF activation by disrupting the 27. Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin J, sur-8-Ras-Raf complex. J Biol Chem 2006;281:927–33. et al. Improved survival with vemurafenib in melanoma with BRAF 10. Rodriguez-Viciana P, Oses-Prieto J, Burlingame A, Fried M, McCor- V600E mutation. N Engl J Med 2011;364:2507–16. mick F. A phosphatase holoenzyme comprised of Shoc2/Sur8 and the 28. Johannessen CM, Boehm JS, Kim SY, Thomas SR, Wardwell L, catalytic subunit of PP1 functions as an M-Ras effector to modulate Johnson LA, et al. COT drives resistance to RAF inhibition through Raf activity. Mol Cell 2006;22:217–30. MAP kinase pathway reactivation. Nature 2010;468:968–72. 11. Takayama S, Sato T, Krajewski S, Kochel K, Irie S, Millan JA, et al. 29. Nazarian R, Shi H, Wang Q, Kong X, Koya RC, Lee H, et al. Melanomas Cloning and functional analysis of BAG-1: a novel Bcl-2-binding acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS protein with anti-cell death activity. Cell 1995;80:279–84. upregulation. Nature 2010;468:973–7. 12. Young JC, Barral JM, Hartl FU. More than folding: localized functions of 30. Villanueva J, Vultur A, Lee JT, Somasundaram R, Fukunaga-Kalabis M, cytosolic chaperones. Trends Biochem Sci 2003;28:541–7. Cipolla AK, et al. Acquired resistance to BRAF inhibitors mediated by a 13. Kabbage M, Dickman MB. The BAG proteins: a ubiquitous family of RAF kinase switch in melanoma can be overcome by cotargeting MEK chaperone regulators. Cell Mol Life Sci 2008;65:1390–402. and IGF-1R/PI3K. Cancer Cell 2010;18:683–95. 14. Wang HG, Takayama S, Rapp UR, Reed JC. Bcl-2 interacting protein, 31. Eggermont AM, Robert C. Melanoma in 2011: a new paradigm tumor BAG-1, binds to and activates the kinase Raf-1. Proc Natl Acad Sci U S for drug development. Nat Rev Clin Oncol 2012;9:74–6. A 1996;93:7063–8. 32. Flaherty KT. BRAF inhibitors and melanoma. Cancer J 2011;17: 15. Gotz R, Wiese S, Takayama S, Camarero GC, Rossoll W, Schweizer U, 505–11. et al. Bag1 is essential for differentiation and survival of hematopoietic 33. Villanueva J, Vultur A, Herlyn M. Resistance to BRAF inhibitors: and neuronal cells. Nat Neurosci 2005;8:1169–78. unraveling mechanisms and future treatment options. Cancer Res 16. Song J, Takeda M, Morimoto RI. Bag1-Hsp70 mediates a physiolog- 2011;71:7137–40. ical stress signalling pathway that regulates Raf-1/ERK and cell 34. Montagut C, Sharma SV, Shioda T, McDermott U, Ulman M, Ulkus LE, growth. Nat Cell Biol 2001;3:276–82. et al. Elevated CRAF as a potential mechanism of acquired resistance 17. Gotz R, Kramer BW, Camarero G, Rapp UR. BAG-1 haplo-insufficiency to BRAF inhibition in melanoma. Cancer Res 2008;68:4853–61. impairs lung tumorigenesis. BMC Cancer 2004;4:85. 35. Emery CM, Vijayendran KG, Zipser MC, Sawyer AM, Niu L, Kim JJ, et al. 18. Wellbrock C, Karasarides M, Marais R. The RAF proteins take centre MEK1 mutations confer resistance to MEK and B-RAF inhibition. Proc stage. Nat Rev Mol Cell Biol 2004;5:875–85. Natl Acad Sci U S A 2009;106:20411–6. 19. Platz A, Egyhazi S, Ringborg U, Hansson J. Human cutaneous 36. Shi H, Moriceau G, Kong X, Lee MK, Lee H, Koya RC, et al. Melanoma melanoma: a review of NRAS and BRAF mutation frequencies in whole-exome sequencing identifies (V600E)B-RAF amplification-medi- relation to histogenetic subclass and body site. Mol Oncol 2008;1: ated acquired B-RAF inhibitor resistance. Nat Commun 2012;3:724. 395–405. 37. Arkin MR, Wells JA. Small-molecule inhibitors of protein–protein inter- 20. Thomas NE. BRAF somatic mutations in malignant melanoma and actions: progressing towards the dream. Nat Rev Drug Discov melanocytic naevi. Melanoma Res 2006;16:97–103. 2004;3:301–17.

www.aacrjournals.org Mol Cancer Ther; 12(11) November 2013 2413

Enthammer et al.

38. Morelli X, Bourgeas R, Roche P. Chemical and structural lessons from 50. Rennefahrt UE, Illert B, Kerkhoff E, Troppmair J, Rapp UR. Constitutive recent successes in protein–protein interaction inhibition (2P2I). Curr JNK activation in NIH 3T3 fibroblasts induces a partially transformed Opin Chem Biol 2011;15:475–81. phenotype. J Biol Chem 2002;277:29510–8. 39. Mullard A. Protein–protein interaction inhibitors get into the groove. 51. Molecular Operating Environment (MOE), 2011.10, Chemical Com- Nat Rev Drug Discov 2012;11:173–5. puting Group Inc., Montreal, QC, Canada. 40. Kelenyi G. On the histochemistry of azo group-free thiazole dyes. 52. Jones G, Willett P, Glen RC, Leach AR, Taylor R. Development and J Histochem Cytochem 1967;15:172–80. validation of a genetic algorithm for flexible docking. J Mol Biol 41. Puchtler H, Sweat Waldrop F, Meloan SN. Application of thiazole dyes 1997;267:727–48. to amyloid under conditions of direct cotton dyeing: correlation of 53. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, histochemical and chemical data. Histochemistry 1983;77:431–45. et al. The Protein Data Bank. Nucleic Acids Res 2000;28:235–42. 42. Sharp A, Crabb SJ, Johnson PW, Hague A, Cutress R, Townsend PA, 54. Sondermann H, Scheufler C, Schneider C, Hohfeld J, Hartl FU, et al. Thioflavin S (NSC71948) interferes with Bcl-2-associated atha- Moarefi I. Structure of a Bag/Hsc70 complex: convergent functional nogene (BAG-1)-mediated protein-protein interactions. J Pharmacol evolution of Hsp70 nucleotide exchange factors. Science 2001;291: Exp Ther 2009;331:680–9. 1553–7. 43. Townsend PA, Cutress RI, Carroll CJ, Lawrence KM, Scarabelli TM, 55. Wolber G, Dornhofer A, Langer T. Efficient overlay of small organic Packham G, et al. BAG-1 proteins protect cardiac myocytes from molecules using 3D pharmacophores. J Computer Aided Mol Des simulated ischemia/reperfusion-induced apoptosis via an alternate 2006;20:773–88. mechanism of cell survival independent of the proteasome. J Biol 56. Wolber G, Langer T. LigandScout: 3-D pharmacophores derived from Chem 2004;279:20723–8. protein-bound ligands and their use as virtual screening filters. J Chem 44. Soule HD, Maloney TM, Wolman SR, Peterson WD, Brenz R, McGrath Inf Model 2004;45:160–9. CM, et al. Isolation and characterization of a spontaneously immor- 57. Wolber G, Seidel T, Bendix F, Langer T. Molecule-pharmacophore talized human breast epithelial cell line, MCF-10. Cancer Res 1990; superpositioning and pattern matching in computational drug design. 50:6075–86. Drug Discov Today 2008;13:23–9. 45. Soule H, Vazquez J, Long A, Albert S, Brennan M. A human cell line 58. Humphrey W, Dalke A, Schulten K. VMD: visual molecular dynamics. from a pleural effusion derived from a breast carcinoma. J Natl Cancer J Mol Graph Model 1996;14:33–8. Inst 1973;51:1409–16. 59. Mansour SJ, Matten WT, Hermann AS, Candia JM, Rong S, Fukasawa 46. Miller FR, Soule HD, Tait L, Pauley RJ, Wolman SR, Dawson PJ, et al. K, et al. Transformation of mammalian cells by constitutively active Xenograft model of progressive human proliferative breast disease. MAP kinase kinase. Science 1994;265:966–70. J Natl Cancer Inst 1993;85:1725–32. 60. Calvo F, Agudo-Ibanez L, Crespo P. The Ras-ERK pathway: under- 47. Marais R, Light Y, Paterson HF, Mason CS, Marshall CJ. Differential standing site-specific signaling provides hope of new anti-tumor regulation of Raf-1, A-Raf, and B-Raf by oncogenic ras and tyrosine therapies. Bioessays 2010;32:412–21. kinases. J Biol Chem 1997;272:4378–83. 61. Cutress RI, Townsend PA, Brimmell M, Bateman AC, Hague A, Pack- 48. Mosmann T. Rapid colorimetric assay for cellular growth and survival: ham G. BAG-1 expression and function in human cancer. Br J Cancer application to proliferation and cytotoxicity assays. J Immunol Meth- 2002;87:834–9. ods 1983;65:55–63. 62. Bardelli A, Longati P, Albero D, Goruppi S, Schneider C, Ponzetto C, 49. Gillies RJ, Didier N, Denton M. Determination of cell number in mono- et al. HGF receptor associates with the anti-apoptotic protein BAG-1 layer cultures. Anal Biochem 1986;159:109–13. and prevents cell death. EMBO J 1996;15:6205–12.

2414 Mol Cancer Ther; 12(11) November 2013 Molecular Cancer Therapeutics

Isolation of a Novel Thioflavin S−Derived Compound That Inhibits BAG-1−Mediated Protein Interactions and Targets BRAF Inhibitor− Resistant Cell Lines

Marion Enthammer, Emmanouil S. Papadakis, Maria Salomé Gachet, et al.

Mol Cancer Ther 2013;12:2400-2414. Published OnlineFirst September 18, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-13-0142

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2013/09/18/1535-7163.MCT-13-0142.DC1

Cited articles This article cites 61 articles, 16 of which you can access for free at: http://mct.aacrjournals.org/content/12/11/2400.full#ref-list-1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/12/11/2400. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.