Novel ATP-Competitive MEK Inhibitor E6201 Is Effective Against Vemurafenib-Resistant Melanoma Harboring the MEK1-C121S Mutation in a Preclinical Model

Published OnlineFirst January 21, 2014; DOI: 10.1158/1535-7163.MCT-13-0667

Molecular Cancer Small Molecule Therapeutics Therapeutics

Novel ATP-Competitive MEK Inhibitor E6201 Is Effective against Vemurafenib-Resistant Melanoma Harboring the MEK1-C121S Mutation in a Preclinical Model

Yusuke Narita1,2, Kiyoshi Okamoto1, Megumi Ikemori Kawada1, Kazuma Takase1, Yukinori Minoshima1, Kotaro Kodama1, Masao Iwata1, Norimasa Miyamoto1,2, and Kohei Sawada1,2

Abstract Many clinical cases of acquired resistance to the BRAF inhibitor vemurafenib have recently been reported. One of the causes of this acquired resistance is the BRAF downstream kinase point mutation MEK1-C121S. This mutation confers resistance to not only vemurafenib, but also to the allosteric MEK inhibitor selumetinib (AZD6244). Here, we investigated the pharmacologic activities and effectiveness of the novel MEK inhibitor E6201 against BRAF (v-raf murine sarcoma viral oncogene homolog B1)-V600E mutant melanoma harboring the MEK1-C121S mutation. A cell-free assay conﬁrmed that E6201 is an ATP-competitive MEK inhibitor, meaning it has a different binding mode with MEK compared with allosteric MEK inhibitors. E6201 is more effective against BRAF-V600E mutant melanoma compared with BRAF wild-type melanoma based on MEK inhibition. We found that the acquired MEK1-C121S mutation in BRAF-V600E mutant melanoma conferred resistance to both vemurafenib and selumetinib but not E6201. The effectiveness of E6201 in this preclinical study is a result of its binding with MEK1 far from the C121S point mutation so the mutation is unable to inﬂuence the MAPK pathway inhibitory activity. These results support further clinical investigation of E6201. Mol Cancer Ther; 13(4); 823–32. 2014 AACR.

Introduction Administration and the European Medicines Agency (10). In certain types of cancer, the activation of kinases is However, most patients with melanomas harboring known to have a role in oncogenicity (1). Therefore, kinase BRAF-V600E progress within 2 to 18 months of the initial inhibitors are an attractive option for use as cancer che- response to vemurafenib (11). Unlike with the other kinase motherapy (2). These drugs confer a survival benefit to inhibitors, no acquired BRAF gatekeeper mutation caused patients with cancer; however, a number of mechanisms by vemurafenib treatment has been reported in a clinical of acquired resistance to kinase inhibitors have been case; however, there are reports of resistance to vemur- reported (3, 4). For example, gatekeeper mutations that afenib mediated through the reactivation of the mitogen- confer resistance to kinase inhibitors frequently occur in activated protein kinase (MAPK) signaling pathway and EGFR (5, 6) and BCR-ABL (7). the activation of the AKT signaling pathway (12–22). Patients with metastatic melanoma have a very poor The MEK1-C121S mutation downstream of BRAF is an prognosis with a median survival time of only 6 to 10 acquired mutation that has recently been found to confer months (8). The BRAF-V600E mutation occurs in approx- resistance to vemurafenib (12). This mutation also confers imately 50% of melanomas, and this mutation is strongly resistance to the allosteric mitogen-activated protein associated with tumorigenicity (8, 9). Recently, a clinical extracellular signal–regulated kinase (MEK) inhibitor trial showed that the BRAF inhibitor vemurafenib con- selumetinib (AZD6244; ref. 12). MEK1-C121S suppresses ferred significant survival benefit in patients with mela- the MAPK pathway inhibitory activity of vemurafenib nomas harboring BRAF-V600E, and vemurafenib has and selumetinib (12). Therefore, to improve the clinical since been approved by both the U.S. Food and Drug outcome of BRAF-V600E melanomas harboring MEK1- C121S, drugs that further inhibit the MAPK pathway are required. Authors' Affiliations: 1Eisai Co., Ltd.; and 2Genomics-Based Drug Dis- covery, Graduate School of Comprehensive Human Sciences, University of E6201 is a novel MEK inhibitor that we developed (23), Tsukuba, Tsukuba, Ibaraki, Japan and a phase I clinical trial of E6201 is ongoing in advanced Note: Supplementary data for this article are available at Molecular Cancer solid tumors (trial registration ID: NCT00794781). The Therapeutics Online (http://mct.aacrjournals.org/). chemical structure of E6201, a derivative of the natural Corresponding Author: Yusuke Narita, Oncology, Eisai Co. Ltd., 5-1-3, product f152A1, is markedly different from the represen- Tokodai, Tsukuba, Ibaraki 300-2635, Japan. Phone: 81-29-847-5291; tative allosteric MEK inhibitors (Fig. 1). A docking sim- Fax: 81-29-847-2037; E-mail: [email protected] ulation showed that f152A1 binds at the ATP binding site doi: 10.1158/1535-7163.MCT-13-0667 of MEK1 (24); therefore, E6201 is also likely to be an ATP- 2014 American Association for Cancer Research. competitive MEK inhibitor with different pharmacologic

www.aacrjournals.org 823

Narita et al.

A B H N HO O O N O O Cl F H H H N N N N

O N N F Br I OH O N O O Figure 1. A, structure of E6201. Selumetinib (AZD6244) Trametinib (GSK1120212) O B, structure of 4 allosteric HN MEK inhibitors. OH H H OH N O N O O Cl O O F E6201 H H N OH N

F I F I F F CI-1040 PD0325901

activities than those of allosteric MEK inhibitors. Further- (1.3 kbp) were purified from agarose gels using a QIAEX II more, E6201 merits further investigation for the treatment Gel Extraction Kit (Qiagen). BP reactions were performed of BRAF-V600E melanomas harboring MEK1-C121S in a total volume of 10 mL containing 2 mL of BP Clonase II because of its different MEK binding mode compared Enzyme Mix, 150 ng pDONR P1-P5r (Invitrogen), and 100 with allosteric MEK inhibitors. ng PCR product. The reaction was incubated for 2 hours at Here, we investigated the use of E6201 as a MEK 25C. After proteinase treatment, the mixture was trans- inhibitor against BRAF-V600E melanoma harboring formed into Escherichia coli TOP10 cells (Invitrogen). Col- MEK1-C121S. We show that E6201 is an ATP-competitive onies that grew on Lysogeny Broth plate (containing 50 MEK inhibitor that has potent growth inhibitory activity mg/mL kanamycin) were picked and the insert was in BRAF-V600E melanoma compared with BRAF wild- sequenced. type (WT) melanoma, although the compound inhibited the MAPK pathway both in BRAF-WT and BRAF-V600E Site-directed mutagenesis melanoma. Furthermore, we demonstrate that E6201 is pENTR_L1-MEK1(C121S)-R5, a MEK1-C121S mutant effective in a preclinical model against BRAF-V600E mel- ENTRY vector, was generated by site-directed mutagen- anoma harboring MEK1-C121S and warrants further clin- esis with a PrimeSTAR Mutagenesis Basal Kit (TaKaRa ical investigation in a larger setting. Bio) according to the manufacturer’s instructions. The primers used were MEK1(C121S)-Fwd., 50-catgagtc- Materials and Methods caactctccgtacatcgtgggc-30 and MEK1(C121S)-Rev., 50- Compounds agagttggactcatgcagaacctgcagctc-30. PCR reactions were Selumetinib and vemurafenib were provided by Wuxi performed in a total volume of 50 mL containing 10 AppTec. mmol/L of each primer, 10 pg of pENTR_L1-MEK1-R5 as the template DNA, and 25 mLof2 PrimeSTAR Max Construction of pENTR_L1-MEK1-R5 ENTRY vector Premix. PCR reactions were incubated at 98C for 1 pF1KE0668, a plasmid harboring the coding regions of minute, followed by 30 cycles of 98C for 10 seconds, the full-length MEK1 gene, was purchased from Kazusa 55C for 15 seconds, and 72C for 50 seconds, with a final DNA Research Institute. The MEK1 coding region was extension period at 72C for 2 minutes. The PCR product amplified using the following primer set: attB1_MEK1- (2.5 mL) was transformed into E. coli TOP10 cells (Invitro- Fwd., 50-ggggacaagtttgtacaaaaaagcaggctgccaccatgcccaag- gen) by using the heat shock method, and recombinants aagaagccgacgcccatcc-30; attB5r_MEK1-Rev., 50-ggggacaa- were selected on Lysogeny Broth agar media supplemen- cttttgtatacaaagttgtttagacgccagcagcatgggttggtgtgctgggc-30. ted with 50 mg/mL kanamycin. Two att-site sequences PCR reactions were performed in a total volume of 50 mL and the MEK1 region were checked by sequencing. containing 10 mmol/L of each primer, 0.5 ng of pF1KE0668 as the template DNA, and 1.25 U PrimeSTAR GXL DNA Construction of pCLxIP_MEK1 (WT or C121S)- polymerase (TaKaRa Bio) according to the manufacturer’s IRES-hmAG vectors (MEK1-internal ribosomal instructions. PCR reactions were incubated at 98C for 1 entry site–dependent Azami-Green protein minute, followed by 30 cycles of 98C for 10 seconds, 60C expression vectors) for 15 seconds, and 68C for 2 minutes, with a final pCLxIP_MEK1-IRES-hmAG vector was constructed by extension period at 68C for 5 minutes. The PCR products using a MultiSite Gateway Pro Kit (Invitrogen) according

824 Mol Cancer Ther; 13(4) April 2014 Molecular Cancer Therapeutics

E6201 Treatment for Acquired Resistance to Vemurafenib

to manufacturer’s instructions. pCLxIP-DEST (Cytomeg- (Illumina). Sequencing data were produced by using a alovirus promoter–based expression vector) and Genome Analyzer II (Illumina) and Standard Cluster Gen- pENTR_L5-IRES-hmAG-L2 [internal ribosomal entry site eration and 36 Cycle Sequencing Kits (Illumina). (IRES)–dependent Azami-Green protein expression vector] were used in this study. These 2 vectors were Western blot analysis created in-house. MultiSite LR reactions were performed Cells were incubated with a serial dilution of the test in a 10-mL total volume containing 10 fmoles pENTR_L1- compounds in complete medium at 37C for 24 hours. MEK1-R5, 10 fmoles pENTR_L5-IRES-hmAG-L2, 20 Cells were lysed with cell lysis buffer (Cell Signaling fmoles pCLxIP-DEST, and 2 mL LR II Clonase Plus (Invi- Technology) containing 1 mmol/L phenylmethylsulfonyl trogen). The reaction was incubated overnight at 25C. fluoride. Cellular debris was removed by centrifugation at After proteinase treatment, the mixture was transformed 15,000 g for 20 minutes at 4C. Western blotting was into E. coli TOP10 cells (Invitrogen) by using the heat performed as described (25). Supernatants containing 10 shock method, and recombinants were selected on Lysog- mg of protein were subjected to SDS-PAGE under reduc- eny Broth agar media supplemented with 100 mg/mL ing conditions. The proteins were then transferred onto ampicillin. The MEK1 and IRES-hmAG regions were polyvinylidene fluoride membranes (Millipore) blocked checked by sequencing. A MEK1-C121S mutant expres- with TBS containing 0.05% Tween-20 and 5% skim milk. sion vector (pCLxIP_MEK1[C121S]-IRES-hmAG) was The membranes were then probed with the following constructed via the same method. These expression vec- antibodies: anti-phospho ERK1/2 (Thr202/Tyr204), tors were designed so that the IRES-dependent Azami- anti-ERK1/2 (Cell Signaling Technology), and anti-cyclin Green gene and puromycin resistance gene were co- D1 (BD Biosciences). Immunoreactive bands were visu- expressed with the MEK1 gene. alized by detecting chemiluminescence with a ChemiDoc XRS gel imaging system (Bio-Rad). Cell lines and cell cultures The human melanoma cell lines HMV-1 and SEKI were Cell-proliferation assay obtained from the Japanese Collection of Research Bior- Cells (2 103 cells/100 mL/well) were seeded in 96-well esources in 1987. The human melanoma cell lines HMCB, culture plates. After overnight incubation at 37C under a CHL-1, MDA-MB435S, and A375 were obtained from the humidified atmosphere containing 5% CO2, various con- American Type Culture Collection in 2008, 2008, 1998, and centrations of each compound were added, and cultured 2010, respectively. The melanoma cell line G-361 was for 3 days. Then, 10 mL of WST-8 reagent (Dojindot) was obtained from DS Pharma Biomedical in 2010. A375/ added to each well, and absorbance was measured at 450 MEK1 WT and A375/MEK1-C121S cells were prepared nm by using an Envision multilabel plate reader (Perkin by stable transfection of MEK1 WT- or C121S-expressing Elmer) and compared with a reference measurement at plasmids into parental A375 cells. HMV-1, SEKI, MDA- 665 nm. MB435S, and A375 cell lines were cultured in RPMI-1640 containing 10% FBS. HMCB and CHL-1 cell lines were In vitro MEK1 and MEK2 kinase assay cultured in Dulbecco’s Modified Eagle Medium contain- An ELISA Assay Kit containing MEK1 and MEK2 ing 10% FBS. G-361 cell line was cultured in McCoy’s recombinant protein was purchased from Carna Bios- medium containing 10% FBS. All cell lines were cultured ciences. The assay was performed according to Carna at 37C under a humidified atmosphere containing 5% Biosciences’ internal protocol. CO2, and authenticated by short tandem repeat analysis. Immunocytochemistry BRAF DNA sequencing A375 cells were seeded (5 105 cells/100 mL/well) in 6- BRAF mutation status was assessed by using next- well plates the day before transfection. The cells were generation sequencing technology. Reverse-transcribed transfected with MEK1 expression vectors using lipofec- cDNAs were prepared from HMV-1, HMCB, CHL-1, tamine 2000 (Invitrogen) according to the manufacturer’s A375, SEKI, MDA-MB435S, and G361 cell lines and then protocol. The following day, the cells were harvested by used as templates for amplification of several genes of trypsinization and seeded (2.5 104 cells/well) in 96-well interest, including BRAF. Because it was difficult to amplify imaging plates (BD falcon). After overnight incubation at full-length BRAF cDNA, 3 pairs of primers that together 37 C under a humidified atmosphere containing 5% CO2, cover the whole BRAF cDNA coding region were used the cells were treated with serial dilutions of the test (sense primer 1: GCCCCGGCTCTCGGTTATAAGATG, compounds for 24 hours. For immunostaining, the cells antisense primer 1: CCGTTCCCCAGAGATTCCAA; sense were fixed with 2% paraformaldehyde for 45 minutes and primer 2: TGCCATTCCGGAGGAGGTGT, antisense permeabilized with 100% MeOH for 10 minutes. After primer 2: GCCCATCAGGAATCTCCCAA; sense primer blocking with 1% bovine serum albumin (BSA)/PBS for 3 3: ATCTGGATCATCCCCTTCCGC, antisense primer 3: hours, the cells were incubated overnight with rabbit anti- CCCGGAACAGAAAGTAAAGCCTCTAG). PCR pro- cyclin D1 (Abcam) in 0.1% BSA/PBS at 4C. After being ducts from each cell line were mixed, and a library was washed with PBS, the cells were incubated with Alexa568- constructed by using a Genomic DNA Sample Prep Kit conjugated anti-rabbit IgG (Invitrogen) and 10 mg/mL

www.aacrjournals.org Mol Cancer Ther; 13(4) April 2014 825

Narita et al.

Hoechst 33342 (Invitrogen) in 0.1% BSA/PBS for 1 hours account for ligand flexibility, as well as the refinement and then washed with PBS. module. The Prime algorithm is implemented in Glide to Image acquisition for each well was performed on an IN account for the flexibility of the receptor. Residues within Cell Analyzer 1000 (GE Healthcare) by using a 20 objec- 10 A˚ of the ligand poses were minimized to form suitable tive lens with 360-, 475-, and 555-nm excitation filters pose conformations at the binding site. Finally, each monitored through 460-, 517-, and 620-nm emission fil- ligand was redocked to its corresponding low-energy ters, respectively. The cyclin D1 expression level of the protein structures and the resulting complexes were vector-transfected Azami-Green–positive cells was deter- ranked according to their GlideScore. The IFD, given in mined by using the IN Cell Developer software (GE kilocalories per mole, was computed based on the Glide- Healthcare). Images collected from 460-nm emission fil- Score and a small fraction of the Prime energy by using the ters were used to define nuclear regions. The Azami- following formula: IFD score ¼ GlideScore þ 0.05 Prime Green expression level and cyclin D1 expression level of energy. each cell were defined as the level of 517- and 620-nm intensity in the nuclear region, respectively. The mean and Results SD background value at 517-nm intensity was determined E6201 is an ATP-competitive MEK1 and MEK2 by analyzing wells that did not contain cells. Cells with inhibitor þ Azami-Green expression levels higher than mean 8 To determine whether E6201 is an ATP-competitive or SD of the background value at 517-nm intensity were -noncompetitive MEK inhibitor, the inhibitory activity of defined as Azami-Green positive. The cyclin D1 expres- E6201 on MEK1 and MEK2 was evaluated by using a cell- sion level of each well was determined as the median free kinase assay with various concentrations of ATP (Fig. cyclin D1 expression level of Azami-Green–positive cells 2). The IC50 values of E6201 in the MEK1 cell-free assay in the well. were 0.020, 0.12, and 1.4 mmol/L at ATP concentrations of m 10, 30, and 100 mol/L, respectively. The IC50 values Protein structure and docking simulations with in the MEK2 cell-free assay were 0.028, 0.067, and 0.70 MEK1 protein mmol/L at ATP concentrations of 10, 30, and 100 mmol/L, Homology modeling of protein structure. The coordi- respectively. The inhibitory activity of E6201 decreased nates for human MEK1 and the allosteric inhibitor G-894 with increasing ATP dose, indicating that E6201 is an complex were obtained from the Research Collaboratory ATP-competitive MEK1 and MEK2 inhibitor. for Structural Bioinformatics Protein Data Bank (PDB entry code: 3V04; ref. 26), and the protein structure was prepared by using the Maestro software package Efficacy of E6201 against BRAF-V600E or -WT (Schroinger).€ Hydrogen atoms were added and a brief melanoma relaxation was performed on each starting structure using BRAF-V600E melanoma is reported to be more sensitive the protein preparation module in the Maestro software. than BRAF-WT melanoma to the MAPK pathway inhibi- These coordinates were used for the selumetinib docking tion caused by BRAF inhibitors or allosteric MEK inhibi- simulation. For the E6201 simulation, a MEK1 model was tors (32, 33). To assess whether E6201 shows selective built by using the Prime homology modeling program (27, growth suppression in BRAF-V600E melanoma, we pre- 28) and the human ERK2 crystal structure (PDB entry pared a melanoma cell-line panel comprising 3 BRAF-WT code: 2E14; ref. 29) coordinates as the template structure. (CHL-1, HMCB, and HMV-1) and 4 BRAF-V600E mela- Induced-fit docking simulation. The induced-fit noma cell lines (G361, MDA-MB435, SEKI, and A375). The docking (IFD; ref. 30) module in the Maestro software BRAF mutation status of these cell lines was confirmed by was used. IFD uses the Glide docking program (31) to DNA sequencing. Cell lines harboring the BRAF-V600E

A B MEK2 MEK1 PTA PTA 120 10 μmol/L 120 10 μmol/L 30 μmol/L 100 30 μmol/L 100 100 μmol/L 100 μmol/L 80 80 Figure 2. MEK1 (A) and MEK2 (B) 60 60 inhibitory activity of E6201 at different concentrations of ATP. 40 40 (% control) of (% control) of 20 20 Substrate phosphorylation Substrate phosphorylation 0 0 0.001 0.01 0.1 1 0.001 0.01 0.1 1 E6201 concentration (μmol/L) E6201 concentration (μmol/L)

826 Mol Cancer Ther; 13(4) April 2014 Molecular Cancer Therapeutics

E6201 Treatment for Acquired Resistance to Vemurafenib

A Vemurafenib Selumetinib E6201

IC50 >1.0 μmol/L IC50 >1.0 μmol/L IC50 >1.0 μmol/L 1.0 1.0 1.0

0.8 0.8 0.8

0.6 0.6 0.6 ( μ mol/L) ( μ mol/L) ( μ mol/L) 50 50 50 0.4 0.4 0.4

0.2 0.2 0.2 Growth IC Growth IC Growth IC 0.0 0.0 0.0 SEKI A375 SEKI SEKI A375 A375 G361 G361 G361 CHL-1 HMCB CHL-1 CHL-1 HMCB HMCB HMV-1 HMV-1 HMV-1 MDA-MB435 MDA-MB435 B MDA-MB435 Vemurafenib Selumetinib E6201 Vemurafenib Selumetinib E6201 0 0.01 0.1 1 0 0.01 0.1 1 0 0.01 0.1 1 0 0.01 0.1 1 0 0.01 0.1 1 0 0.01 0.1 1 (μmol/L)

pERK

ERK

Cyclin D1

β-Actin

HMV-1 (BRAF-WT) A375 (BRAF-V600E)

Figure 3. Effect of vemurafenib (BRAF inhibitor), selumetinib (allosteric MEK inhibitor), and E6201 (ATP-competitive MEK inhibitor) on BRAF-WT and -V600E melanoma cell lines. BRAF mutation status was conﬁrmed by using DNA sequencing. A, growth inhibitory activity of each compound. White columns represent the result in BRAF-WT cell lines and black columns represent the result in BRAF-V600E cell lines. The growth IC50 values of each compound in BRAF-WT melanoma cell lines were all above 1 mmol/L. B, MAPK pathway inhibitory activity of the test compounds. mutation were more sensitive to vemurafenib and selu- noma. These results indicate that the growth inhibitory metinib compared with BRAF-WT cell lines (Fig. 3A). activity of E6201 in melanoma cell lines harboring the E6201 also showed potent growth inhibitory activity in BRAF-V600E mutation is mediated via MEK inhibition. BRAF-V600E melanoma cell lines compared with BRAF- WT cell lines. E6201 effectiveness on cells transiently or Next, we examined the MAPK pathway inhibitory constitutively expressing MEK1-C121S activity of E6201 in a BRAF-V600E mutant (A375 and MEK1-C121S point mutation confers resistance to both G361) and BRAF-WT melanoma (HMV-1) cell line (Fig. vemurafenib and selumetinib (12). E6201 is an ATP-com- 3B and Supplementary Fig. S1). Vemurafenib inhibited petitive MEK inhibitor that binds to MEK1 or MEK2 ERK phosphorylation in the BRAF-V600E mutant but not differently from allosteric MEK inhibitors. E6201 may in the BRAF-WT melanoma (Fig. 3B), whereas selumetinib therefore be effective against BRAF-V600E melanomas inhibited ERK phosphorylation in both the BRAF-V600E harboring MEK1-C121S. mutant and the BRAF-WT melanoma. E6201 also inhib- To assess whether MEK1-C121S confers resistance to ited ERK phosphorylation in both the BRAF-V600E E6201 in melanoma cells, we transiently transfected A375 mutant and the BRAF-WT melanoma (Fig. 3B and Sup- human melanoma cells with a MEK1-C121S–expressing plementary Fig. S1). E6201 inhibited ERK phosphoryla- vector. A375 was used because this cell line is reported to tion at a lower dose in the BRAF-V600E mutant than in the harbor the BRAF-V600E mutation and has frequently BRAF-WT melanoma, whereas selumetinib inhibited been used in similar experiments (12, 34). The presence ERK phosphorylation in the BRAF-V600E mutant at of the BRAF-V600E mutation in A375 was conﬁrmed approximately the same dose as in the BRAF-WT mela- by DNA sequencing. After treatment with the test

www.aacrjournals.org Mol Cancer Ther; 13(4) April 2014 827

Narita et al.

A DMSO Vemurafenib Selumetinib E6201

Hoechst

Azami-Green

Cyclin D1

Azami-Green + Cyclin D1

Vemurafenib Selumetinib E6201 Nontransfection 160 160 Nontransfection 160 MEK1-C121S Nontransfection 140 140 MEK1-C121S 140 MEK1-C121S 120 120 120 100 100 100 80 80 80 60 60 60 (% of control) (% of (% of control) (% of 40 40 control) (% of 40

20 Cyclin D1 expression

20 Cyclin D1 expression 20 Cyclin D1 expression 0 0 0 0.01 0.1 1 10 100 0.01 0.1 1 10 0.01 0.1 1 10 Concentration Concentration Concentration (μmol/L) (μmol/L) (μmol/L)

Figure 4. Effect of vemurafenib, selumetinib, or E6201 on cyclin D1 expression in exogenous MEK1-C121S–positive or -negative A375 cells. A, cells were treated with 3.3 mmol/L vemurafenib, 1.1 mmol/L selumetinib, or 1.1 mmol/L E6201. Hoechst is a total cell stain. Azami-Green–positive cells represent exogenous MEK1-C121S–expressing cells. Scale bar, 100 mm. B, quantiﬁcation of cyclin D1 expression in exogenous MEK1-C121S–positive or -negative cells after treatment with the test compounds.

compounds, transfected cells were immunocytochemical- WT/C121S–negative and -positive) cells separately by ly stained for cyclin D1, which is a marker of the cytostatic using an IN Cell Analyzer high-content analysis system. effect of MAPK inhibitors (32, 35, 36). Only about 10% of cyclin D1 expression decreased after treatment with 3.3 cells were transfected with expression vectors in our mmol/L vemurafenib or 1.1 mmol/L selumetinib in almost transfection protocol, so to be able to identify transfected all of the Azami-Green–negative cells, but not in all of the cells, we designed expression vectors for the co-expres- Azami-Green–positive cells (Fig. 4A). However, cyclin D1 sion of the IRES-dependent Azami-Green and MEK1 expression decreased after treatment with 1.1 mmol/L genes. We quantiﬁed the expression of cyclin D1 in E6201 in both the Azami-Green–positive and -negative Azami-Green–negative and -positive (exogenous MEK1- cells. Quantitatively, the cyclin D1 expression inhibitory

828 Mol Cancer Ther; 13(4) April 2014 Molecular Cancer Therapeutics

E6201 Treatment for Acquired Resistance to Vemurafenib

A Vemurafenib Selumetinib E6201

120 MEK1-WT 120 MEK1-WT 120 MEK1-WT MEK1-C121S MEK1-C121S 100 100 100 MEK1-C121S 80 80 80

60 60 60 40 40

Proliferation 40 Proliferation Proliferation (% of control) (% of (% of control) (% of (% control) of 20 20 20

0 0 0 0.001 0.01 0.1 1 10 100 0.001 0.01 0.1 1 10 100 0.001 0.01 0.1 1 10 100 Concentration (μmol/L) Concentration (μmol/L) Concentration (μmol/L) B

MEK1-WT Vemurafenib Selumetinb E6201

0 0.02 0.2 2 0 0.02 0.2 2 0 0.02 0.2 2 (μmol/L)

pERK

ERK

Cyclin D1

β-Actin

MEK1-C121S Vemurafenib Selumetinb E6201 0 0.02 0.2 2 0 0.02 0.2 2 0 0.02 0.2 2 (μmol/L)

pERK

ERK

Cyclin D1

β-Actin

Figure 5. Effect of vemurafenib, selumetinib, or E6201 in MEK1-WT or -C121S transfectants. A, growth inhibitory activity of the test compounds. B, MAPK pathway inhibitory activity of each compound. activities of vemurafenib and selumetinib in Azami- G361 (Supplementary Fig. S3). In A375 and G361, apopto- Green–positive cells (exogenous MEK1-C121S–positive sis was not inducted by MAPK inhibition (Supplementary cells) were remarkably lower compared with that in Fig. S4). Azami-Green–negative cells (exogenous MEK1-C121S- Next, to conﬁrm the result of the transient transfec- negative cells; Fig. 4B). The cyclin D1 expression inhibitory tion experiments about growth inhibition, stable MEK1- activity of vemurafenib and selumetinib in Azami-Green– WT or -C121S transfectants were established by using positive cells (exogenous MEK1-WT–positive cells) did the A375 cell line. The growth inhibitory activities of not change compared with that in Azami-Green–negative vemurafenib and selumetinib in the MEK1-C121S trans- cells (exogenous MEK1-WT–positive cells; Supplementa- fectants were approximately 10 times weaker than in the ry Fig. S2). In contrast, the cyclin D1 expression inhibitory MEK1-WTtransfectants(Fig.5A).However,thatof activity of E6201 in Azami-Green–positive (exogenous E6201 in the MEK1-C121S transfectant was that same MEK1-WT or C121S-positive) cells barely changed com- as that in the MEK1-WT transfectants (Fig. 5A). These pared with Azami-Green–negative (exogenous MEK1- results indicate that the MEK1-C121S mutation confers WT- or C121S-negative) cells (Fig. 4 and Supplementary resistance to both vemurafenib and selumetinib but not Fig. S2). Moreover, the same result was obtained by using E6201.

www.aacrjournals.org Mol Cancer Ther; 13(4) April 2014 829

Narita et al.

Mechanism of acquired vemurafenib and selumetinib mutations are unable to affect its inhibitory activity. These resistance and effectiveness of E6201 against MEK1- results indicate that the different actions of selumetinib C121S melanoma and E6201 in BRAF-V600E melanoma result from the To conﬁrm the mechanism of vemurafenib and selu- different binding modes to MEK 1. metinib resistance and the effectiveness of E6201 against MEK1-C121S mutant melanoma, ERK phosphorylation inhibitory activity was assessed in A375 MEK1-WT and Discussion -C121S transfectants (Fig. 5B). The ERK phosphorylation On the basis of its MEK inhibitory activity, E6201 was inhibitory activities of vemurafenib and selumetinib in the shown here to be an ATP-competitive MEK inhibitor that MEK1-C121S transfectant were approximately 10 times is effective against BRAF-V600E melanoma in a preclinical lower than those in the MEK1-WT transfectants. In con- setting. Furthermore, we showed that E6201 is effective in trast, the inhibitory activity of E6201 in the MEK1-C121S a preclinical model against BRAF-V600E melanoma har- transfectants was the same as that in the MEK1-WT boring the MEK1-C121S mutation. Almost all of the MEK transfectants. These results indicate that vemurafenib and inhibitors previously reported are allosteric inhibitors, not selumetinib resistance and the effectiveness of E6201 ATP-competitive inhibitors (37). Our docking simulation against MEK1-C121S mutant melanoma is caused by showed that E6201 and selumetinib bind to different sites inhibition of the MAPK pathway. when they dock with MEK. These results suggest that the Both selumetinib and E6201 are MEK inhibitors; how- inhibitory mode of action of E6201 is different from that of ever, the MEK1-C121S mutation confers resistance only to allosteric MEK inhibitors and that E6201 merits further selumetinib. To elucidate why this difference occurs, we investigation in a clinical setting against both MEK-WT simulated the docking process between selumetinib or melanomas and melanomas harboring MEK with muta- E6201 and MEK1. The most energy-stable models are tions at the allosteric site. presented in Fig. 6. Selumetinib binds to a hydrophobic The MEK1-C121S mutation, which confers resistance to pocket adjacent to the ATP binding site, like other allo- vemurafenib, increases MEK activity independently from steric inhibitors (34). This hydrophobic pocket includes BRAF and also confers resistance to the allosteric MEK residues from both a-helix C and the activation loop. The inhibitor selumetinib (12). Our docking simulation binding of allosteric inhibitors within this pocket prevents showed that the MEK1-C121S mutation induced by the structural reorganization of a-helix C and other vemurafenib treatment is located where selumetinib motifs, which generates a catalytically active MEK1 con- binds with MEK1, but is far from where E6201 binds with formation. a-Helix C contains Cys_121, so Cys_121 muta- MEK1. This may account for the selumetinib resistance tions may cause resistance either by direct interference or in BRAF-V600E melanoma harboring the MEK1-C121S by alteration of the C helix conformation. E6201, however, mutation. In contrast, E6201 suppressed ERK phosphor- binds at the ATP binding site, meaning that Cys_121 ylation and inhibited cell growth equivalently in

A Top view Side view

Cys_121

ATP Cys_121 Selumetinib

Selumetinib Figure 6. A, most energy-stable ATP Selumetinib complex models. Selumetinib binds to a hydrophobic pocket adjacent to the ATP binding site. B, most energy-stable B TopTopview View Side view complex model. E6201 binds at the ATP binding site.

Cys_121 Cys_121

E6201

E6201 E6201

830 Mol Cancer Ther; 13(4) April 2014 Molecular Cancer Therapeutics

E6201 Treatment for Acquired Resistance to Vemurafenib

BRAF-V600E melanoma harboring MEK1-C121S and in E6201 against BRAF-WT and BRAF-V600E may lead to a WT melanoma cells. Thus, this is the first paper showing broader therapeutic window because of its weak activity that E6201 is potentially effective against melanomas with in BRAF-WT cells. acquired resistance to vemurafenib associated with the In conclusion, this preclinical study might lead to sup- MEK1-C121S mutation. Further investigation in a clinical port the clinical development of E6201, an ATP-compet- setting in a larger trial is warranted. itive MEK inhibitor, in BRAF inhibitor or allosteric MEK Our preclinical data suggest that E6201 warrants fur- inhibitor resistant cancer. ther study as a novel and potentially favorable therapeutic approach for melanoma. One option for using E6201 in Disclosure of Potential Conflicts of Interest the clinical setting is as second-line monotherapy after No potential conflicts of interest were disclosed. first-line treatment with vemurafenib in BRAF-V600E melanoma. For this approach, companion diagnostics to Authors' Contributions detect the MEK1-C121S mutation status are needed for Conception and design: Y. Narita, K. Okamoto Development of methodology: Y. Narita, K. Okamoto appropriate patient stratification. Acquisition of data (provided animals, acquired and managed patients, Another option for using E6201 in the clinical setting provided facilities, etc.): Y. Narita, K. Okamoto, Y. Minoshima is in combination with vemurafenib as first-line therapy Analysis and interpretation of data (e.g., statistical analysis, biostatis- tics, computational analysis): Y. Narita, K. Okamoto, M.I. Kawada in metastatic melanomas harboring the BRAF- Writing, review, and/or revision of the manuscript: Y. Narita, K. Oka- V600E mutation. A combination therapy clinical trial moto, M.I. Kawada, K. Takase, Y. Minoshima, K. Kodama, N. Miyamoto Administrative, technical, or material support (i.e., reporting or orga- in melanoma using a BRAF inhibitor together with an nizing data, constructing databases): M.I. Kawada, K. Takase, M. Iwata allosteric MEK inhibitor is ongoing (trial registration ID: Study supervision: Y. Narita, M. Iwata, K. Sawada NCT01072175, NCT01271803); however, this combination may not be effective upon acquisition of the MEK1- Acknowledgments C121S mutation because the mutation confers resistance We are grateful to T. Owa, K. Nomoto, T. Uenaka, and A. Tsuruoka for to both BRAF inhibitors and allosteric MEK inhibitors. coordinating the research team. The combination of vemurafenib and E6201 may lead to a prolonged clinical response compared with vemur- Grant Support All authors received financial support by Eisai Co. Ltd. afenib monotherapy by inhibiting acquisition of the The costs of publication of this article were defrayed in part by the MEK1-C121S mutation. payment of page charges. This article must therefore be hereby marked We show here that the inhibitory activity of E6201 was advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. higher in BRAF-V600E melanoma than in BRAF-WT melanoma, whereas that of selumetinib in BRAF-V600E was Received August 13, 2013; revised January 15, 2014; accepted January 16, the same as in BRAF-WT. This inhibitory selectivity of 2014; published OnlineFirst January 21, 2014.

References 1. Tsatsanis C, Spandidos DA. Oncogenic kinase signaling in human 11. Tuma RS. Getting around PLX4032: studies turn up unusual mechan- neoplasms. Ann N Y Acad Sci 2004;1028:168–75. isms of resistance to melanoma drug. J Natl Cancer Inst 2011;103: 2. Grant SK. Therapeutic protein kinase inhibitors. Cell Mol Life Sci 170–1, 117. 2009;66:1163–77. 12. Wagle N, Emery C, Berger MF, Davis MJ, Sawyer A, Pochanard P, et al. 3. Yauch RL, Settleman J. Recent advances in pathway-targeted cancer Dissecting therapeutic resistance to RAF inhibition in melanoma by drug therapies emerging from cancer genome analysis. Curr Opin tumor genomic profiling. J Clin Oncol 2011;29:3085–96. Genet Dev 2012;22:45–9. 13. Shi H, Moriceau G, Kong X, Lee MK, Lee H, Koya RC, et al. Melanoma 4. Barouch-Bentov R, Sauer K. Mechanisms of drug resistance in whole-exome sequencing identifies (V600E)B-RAF amplification- kinases. Expert Opin Investig Drugs 2011;20:153–208. mediated acquired B-RAF inhibitor resistance. Nat Commun 2012;3: 5. Engelman JA, Settleman J. Acquired resistance to tyrosine kinase 724. inhibitors during cancer therapy. Curr Opin Genet Dev 2008;18:73–9. 14. Nazarian R, Shi H, Wang Q, Kong X, Koya RC, Lee H, et al. Melanomas 6. Hammerman PS, Janne PA, Johnson BE. Resistance to epidermal acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS growth factor receptor tyrosine kinase inhibitors in non-small cell lung upregulation. Nature 2010;468:973–7. cancer. Clin Cancer Res 2009;15:7502–9. 15. Johannessen CM, Boehm JS, Kim SY, Thomas SR, Wardwell L, 7. Milojkovic D, Apperley J. Mechanisms of resistance to imatinib and Johnson LA, et al. COT drives resistance to RAF inhibition through second-generation tyrosine inhibitors in chronic myeloid leukemia. MAP kinase pathway reactivation. Nature 2010;468:968–72. Clin Cancer Res 2009;15:7519–27. 16. Villanueva J, Vultur A, Lee JT, Somasundaram R, Fukunaga-Kalabis M, 8. Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin J, Cipolla AK, et al. Acquired resistance to BRAF inhibitors mediated by a et al. Improved survival with vemurafenib in melanoma with BRAF RAF kinase switch in melanoma can be overcome by cotargeting MEK V600E mutation. N Engl J Med 2011;364:2507–16. and IGF-1R/PI3K. Cancer Cell 2010;18:683–95. 9. Hoeflich KP, Gray DC, Eby MT, Tien JY, Wong L, Bower J, et al. 17. Poulikakos PI, Persaud Y, Janakiraman M, Kong X, Ng C, Moriceau G, Oncogenic BRAF is required for tumor growth and maintenance in et al. RAF inhibitor resistance is mediated by dimerization of aberrantly melanoma models. Cancer Res. 2006;66:999–1006. spliced BRAF(V600E). Nature 2011;480:387–90. 10. Ascierto PA, Kirkwood JM, Grob JJ, Simeone E, Grimaldi AM, Maio M, 18. Basile KJ, Abel EV, Aplin AE. Adaptive upregulation of FOXD3 and et al. The role of BRAF V600 mutation in melanoma. J Transl Med resistance to PLX4032/4720-induced cell death in mutant B-RAF 2012;10:85. melanoma cells. Oncogene 2012;31:2471–9.

www.aacrjournals.org Mol Cancer Ther; 13(4) April 2014 831

Narita et al.

19. Yadav V, Zhang X, Liu J, Estrem S, Li S, Gong XQ, et al. Reac- 27. Jacobson MP, Friesner RA, Xiang Z, Honig B. On the role of the crystal tivation of mitogen-activated protein kinase (MAPK) pathway by environment in determining protein side-chain conformations. J Mol FGF receptor 3 (FGFR3)/Ras mediates resistance to vemurafenib in Biol 2002;320:597–608. human B-RAF V600E mutant melanoma. J Biol Chem 2012;287: 28. Jacobson MP, Pincus DL, Rapp CS, Day TJ, Honig B, Shaw DE, et al. A 28087–98. hierarchical approach to all-atom protein loop prediction. Proteins 20. Su F, Bradley WD, Wang Q, Yang H, Xu L, Higgins B, et al. Resistance 2004;55:351–67. to selective BRAF inhibition can be mediated by modest upstream 29. Ohori M, Kinoshita T, Yoshimura S, Warizaya M, Nakajima H, Miyake H. pathway activation. Cancer Res 2012;72:969–78. Role of a cysteine residue in the active site of ERK and the MAPKK 21. Straussman R, Morikawa T, Shee K, Barzily-Rokni M, Qian ZR, Du J, family. Biochem Biophys Res Commun 2007;353:633–7. et al. Tumour micro-environment elicits innate resistance to RAF 30. Sherman W, Day T, Jacobson MP, Friesner RA, Farid R. Novel pro- inhibitors through HGF secretion. Nature 2012;487:500–4. cedure for modeling ligand/receptor induced ﬁt effects. J Med Chem 22. Wilson TR, Fridlyand J, Yan Y, Penuel E, Burton L, Chan E, et al. 2006;49:534–53. Widespread potential for growth-factor-driven resistance to antican- 31. Friesner RA, Banks JL, Murphy RB, Halgren TA, Klicic JJ, Mainz DT, cer kinase inhibitors. Nature 2012;487:505–9. et al. Glide: a new approach for rapid, accurate docking and scoring. 1. 23. Goto M, Chow J, Muramoto K, Chiba K, Yamamoto S, Fujita M, et al. Method and assessment of docking accuracy. J Med Chem 2004;47: E6201 [(3S,4R,5Z,8S,9S,11E)-14-(ethylamino)-8, 9,16-trihydroxy- 1739–49. 3,4-dimethyl-3,4,9,19-tetrahydro-1H-2-benzoxacyclotetradecine- 32. Solit DB, Garraway LA, Pratilas CA, Sawai A, Getz G, Basso A, et al. 1,7 (8H)-dione], a novel kinase inhibitor of mitogen-activated protein BRAF mutation predicts sensitivity to MEK inhibition. Nature 2006; kinase/extracellular signal-regulated kinase kinase (MEK)-1 and 439:358–62. MEK kinase-1: in vitro characterization of its anti-inﬂammatory and 33. Tsai J, Lee JT, Wang W, Zhang J, Cho H, Mamo S, et al. Discovery of a antihyperproliferative activities. J Pharmacol Exp Ther 2009;331: selective inhibitor of oncogenic B-Raf kinase with potent antimela- 485–95. noma activity. Proc Natl Acad Sci U S A 2008;105:3041–6. 24. Ikemori-Kawada M, Inoue A, Goto M, Wang YJ, Kawakami Y. Docking 34. Emery CM, Vijayendran KG, Zipser MC, Sawyer AM, Niu L, Kim JJ, et al. simulation study and kinase selectivity of f152A1 and its analogs. MEK1 mutations confer resistance to MEK and B-RAF inhibition. Proc J Chem Info Model 2012;52:2059–68. Natl Acad Sci U S A 2009;106:20411–6. 25. Matsui J, Funahashi Y, Uenaka T, Watanabe T, Tsuruoka A, Asada 35. Greger JG, Eastman SD, Zhang V, Bleam MR, Hughes AM, Smitheman M. Multi-kinase inhibitor E7080 suppresses lymph node and KN, et al. Combinations of BRAF, MEK, and PI3K/mTOR inhibitors lung metastases of human mammary breast tumor MDA-MB- overcome acquired resistance to the BRAF inhibitor GSK2118436 231 via inhibition of vascular endothelial growth factor-receptor dabrafenib, mediated by NRAS or MEK mutations. Mol Cancer Ther (VEGF-R) 2 and VEGF-R3 kinase. Clin Cancer Res. 2008;14: 2012;11:909–20. 5459–65. 36. Paraiso KH, Fedorenko IV, Cantini LP, Munko AC, Hall M, Sondak VK, 26. Heald RA, Jackson P, Savy P, Jones M, Gancia E, Burton B, et al. et al. Recovery of phospho-ERK activity allows melanoma cells to Discovery of novel allosteric mitogen-activated protein kinase kinase escape from BRAF inhibitor therapy. Br J Cancer 2010;102:1724–30. (MEK) 1,2 inhibitors possessing bidentate Ser212 interactions. J Med 37. Trujillo JI. MEK inhibitors: a patent review 2008–2010. Expert Opin Ther Chem 2012;55:4594–604. Pat 2011;21:1045–69.

832 Mol Cancer Ther; 13(4) April 2014 Molecular Cancer Therapeutics

Novel ATP-Competitive MEK Inhibitor E6201 Is Effective against Vemurafenib-Resistant Melanoma Harboring the MEK1-C121S Mutation in a Preclinical Model

Yusuke Narita, Kiyoshi Okamoto, Megumi Ikemori Kawada, et al.

Mol Cancer Ther 2014;13:823-832. Published OnlineFirst January 21, 2014.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-13-0667

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2014/01/22/1535-7163.MCT-13-0667.DC1

Cited articles This article cites 37 articles, 11 of which you can access for free at: http://mct.aacrjournals.org/content/13/4/823.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/13/4/823.full#related-urls

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/13/4/823. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.