|Home | About | Journals | Submit | Contact Us | Français|
The BRAF oncoprotein is mutated in about half of malignant melanomas and other cancers, and a kinase activating single valine to glutamate substitution at residue 600 (BRAFV600E) accounts for over 90% of BRAF-mediated cancers. Several BRAFV600E inhibitors have been developed, although they harbor some liabilities, thus motivating the development of other BRAFV600E inhibitor options. We report here the use of an ELISA based high-throughput screen to identify a family of related quinolol/naphthol compounds that preferentially inhibit BRAFV600E over BRAFWT and other kinases. We also report the X-ray crystal structure of a BRAF/quinolol complex revealing the mode of inhibition, employ structure-based medicinal chemistry efforts to prepare naphthol analogs that inhibit BRAFV600E in vitro with IC50 values in the 80–200 nM under saturating ATP concentrations, and demonstrate that these compounds inhibit MAPK signaling in melanoma cells. Prospects for improving the potency and selectivity of these inhibitors are discussed.
RAF family kinases are central players in the highly conserved MAPK signaling pathway (RAS-RAF-MEK-ERK), which relay signals from the extracellular space through receptor tyrosine kinases (RTKs) to the nucleus to promote the expression of genes involved in cell proliferation and survival. RAF kinases function by specifically phosphorylating MEK1/2 on the kinase activation loop, leading to the subsequent activation of ERK1/2 1. The three isoforms of the RAF kinases, ARAF, BRAF, and CRAF, share a common three-domain structure containing the conserved regions CR1, CR2 and CR3. The regulation of RAF kinase activity is complex and involves a variety of different kinases and scaffolding proteins 1. However, compared to ARAF and CRAF, BRAF requires fewer regulation events for its activation. Due to this property, BRAF has significantly higher basal activity than the other RAF family isoforms and is also found to be a major activator of MEK1/2 2, 3.
BRAF is an important oncogene that is mutated in about a half of malignant melanomas and at a lower frequency in a wide range of other human cancers, such as thyroid, colon, ovarian, lung and breast cancers 4, 5. Oncogenic mutations occur within the kinase domain, with a single valine to glutamate substitution at residues 600 (BRAFV600E) accounting for over 90% of these mutations 1, 4, 6, 7. Because of these findings, BRAF, and BRAFV600E in particular, has emerged as an attractive anti-cancer drug target. A number of inhibitors against BRAF have been developed so far and more are at various stages of preclinical and clinical development 8–10. For example, vemurafenib (PLX4032), an azaindole compound and orally available ATP competitive BRAF inhibitor that shows selectivity for BRAFV600E, has received FDA approval for the treatment of late stage metastatic melanoma 12. Encouragingly, in phase III clinical trials, PLX4032 produced 2 complete responses and 24 partial responses out of 32 patients, extending life in many cases by more than 6 months, prior to the eventual developed drug resistance through reactivation of the MAPK pathway or through the activation of alternative compensatory pathways, involving receptor tyrosine kinases (RTKs), PI3K-AKT and other pathways 13, 14. However, about half of the patients had the drug dose reduced, and nearly two-thirds had to have their treatments temporarily stopped, because of side effects 9, 10. In addition, approximately one-quarter of patients developed cancerous or precancerous non-melanoma skin lesions 9, 10. Given these limitations of vemurafenib, and some other drugs that are being evaluated to treat metastatic melanoma, it would be useful to have additional BRAFV600E inhibitor drug options for some patients.
Here we report the development of an ELISA-based high-throughput assay to screen a combined diverse library of over 30,000 organic compounds for BRAFV600E inhibition. This screen, the structure determination of BRAF bound to one of the identified inhibitors, and the follow-up structure-based medicinal chemistry efforts resulted in the identification of a family of related compounds containing a quinolol or naphthol backbone that selectively inhibit BRAFV600E over BRAFWT in vitro, display IC50 values in the 80–200 nM range under saturation ATP concentrations, and inhibit MAPK signaling in melanoma cells.
In order to screen for BRAF inhibitors in a high-throughput format, we employed an Enzyme-Linked Immunosorbent Assay (ELISA)-based system that was previously used by others 15 and us 16. The details of the assay were essentially as we described previously 16. Briefly, we carried out the screen in a 96-well microtiter plate containing covalently immobilized glutathione to enable the capture of glutathione-S-transferase (GST) fusion protein (Pierce Biotechnology) linked to the full-length MEK protein. A mouse monoclonal antibody recognizing BRAF-phosphorylated residues on MEK is then added to the microtiter plate to bind the immobilized and phosphorylated GST-MEK. Antibody molecules that were non-specifically bound are then washed away and horseradish peroxidase (HRP) linked IgG is added to bind to the primary antibody. The amount of HRP IgG binding was then quantified by measuring the chemiluminescence generated by the mixing of the HRP substrate with specifically bound HRP IgG. As shown in Figure 1A, a Z′ factor of 0.82 was obtained indicating that this BRAF ELISA-based assay was effective for high throughput screening.
We used the ELISA-based HTS system to screen a combined library of 31,976 diverse compounds (at a compound concentration of 5 uM, a BRAFV600E concentration of 140 nM and a saturating ATP concentration of 100 uM) (Figures 1B and 1C). From this library, the 23 top scoring compounds deemed to have drug-like properties were purchased directly from their corresponding vendors and were confirmed in a repeat assay to inhibit BRAFV600E activity to more than 90% at an inhibitor concentration of 5 μM (Figure S1A and Compounds 1 – 23 in Figure S1B). Encouragingly, a known CRAF inhibitor was identified in three independent libraries (15 in Figure S1B) and scored among the top 30 inhibitor hits from this experiment, confirming the efficacy of this assay system.
Upon close examination of the representative scaffolds with a rough filtration according to Lipinski’s rule of five 17, 23 compounds were determined to display drug-like structures (Compounds 1 – 23 in Figure S1B). Notably, seven of the compounds contained a quinolol, naphthol or related substructure (compounds 1, 2, 3, 6, 7, 16 and 17 in Figure S1B). Given that this was the largest set of related compounds that we identified in our screen and that these compounds were not previously identified as kinase inhibitors, we focused on their further analysis.
Five of these quinolol/naphthol analogues, in addition to two other closely related compounds (compounds 4 and 5) (Figure 2A), were further characterized by measuring dose response curves for their inhibition against both BRAFWT and BRAFV600E at a protein concentration of 30 nM and inhibitor concentrations ranging from 10 uM to 5 nM to approximate their IC50 values (Figure 2A). This analysis revealed that the most potent inhibitor from these series, 2, inhibited BRAFV600E with an IC50 value of about 100 nM and with about a 3-fold selectivity for BRAFV600E over BRAFWT. Each of the other related inhibitors also displayed strong inhibition towards BRAFV600E with IC50 values within the sub-micro-molar range. All but 5 showed a 2–5 fold selectivity for BRAFV600E over BRAFWT.
In order to understand the mode of action of this family of BRAF inhibitors, co-crystallization of the BRAF kinase domain (BRAF-KD, residues 433–726) with several of the inhibitors were attempted. Since the MALDI-TOF mass spectrum data of BRAF-KD expressed and purified from Sf9 insect cells indicated that the protein is extensively phosphorylated, lambda protein phosphatase (λPP) treatment was used to generate homogeneous hypophosphorylated protein samples for co-crystallization. Extensive co-crystallization efforts using the microbatch method produced co-crystals with only 1 with the wild-type BRAF kinase domain. Crystals of the complex were small (300 μm × 30 μm) producing only low resolution data using a home X-ray source, but data to about 2.5 Å resolution could be obtained using a 10 micron mini-beam at the Advanced Photon Synchrotron source. The BRAF-KD/1 structure was determined by molecular replacement using the unliganded BRAF-KD/sorafenib crystal structure (PDB accession code: 1UWH) as a search model. The BRAF-KD/1 crystals contained two copies of the BRAF-KD/1 complex in one asymmetric unit cell. Electron density corresponding to 1 inhibitor was visible in the ATP binding pockets of both molecules in the asymmetric unit cell. The structure was refined with strict NCS symmetry imposed with the inhibitor modeled only at the later stages of refinement. The final structure was determined to 2.55 Å resolution to an Rwork/Rfree of 0.2203/0.2659 with good geometry (Table 1).
The BRAF-KD/1 structure revealed that the inhibitor binds within the ATP binding cleft between the N-lobe and C-lobe of the kinase domain and an overlay with the structure of PKA in complex with ATP revealed significant overlap of 1 with both the adenine and the ribose moieties of the ATP molecule (Figure 3A). The activation loop of BRAF bound to 1 takes on the active conformation that is observed with BRAF in complex with the PLX4720 (Figure 3B) 18 and CS292 19 inhibitors as opposed to the inactive conformation that is observed with BRAF bound to sorafenib (Figure 3C).
The R-stereoisomer of 1 appears to be bound to the BRAF ATP binding pocket. The quinoline ring is stacked between the N- and C-lobes and against the kinase hinge region with the chloride atom pointing towards the DFG loop with the furan pointing towards the P-loop and the pyridine pointing in the opposite direction towards the D helix (Figure 4A). Specifically, the quinolinol moiety of 1 is intercalated into the space between residues Trp531 and Phe583 forming π-π stacking interactions (Figure 4B). In addition, the nitrogen atom of the quinoline ring of 1 forms hydrogen-bonding interactions with the hinge region of the N- and C-lobes, potentially through the bridging of a water molecule to the main chain carbonyl of residue Cys532. There are also extensive van der Waals contacts between 1 and other residues in the protein pocket such as residues Ile463, Val471, Ala481, Gly464, Thr529, Gln530, Ser535, Ser536, His539, Asn580, Asn581 and Gly593. The furan group is pointing to the extension of the P-loop and the pyridine group is in proximity to His539 of the αD helix. Similar to other BRAF/inhibitor structures with the protein in the active conformation, an 11 Å long “Raf specificity pocket” that is defined by the DFG motif sequence and the αC helix is lined by residues Thr529 (the gatekeeper residue), Leu505, Leu514, Gly593, Asp594, and Phe595 18 (Figure 4C). Taken together, structural evidence reveals extensive and specific interactions between 1 and the ATP binding pocket of the BRAF kinase domain, establishing 1 as a Type I inhibitor, binding to the ATP pocket in the active conformation, confirming its inhibition properties against both BRAFWT and BRAFV600E.
Of the family of quinolol/naphthol BRAF inhibitors that were identified in the screen, we carried out a structure-activity relationship (SAR) around 2 (Figure 2A) because it was the most potent BRAFV600E inhibitor identified in our screen and it appeared to be most tractable for medicinal chemistry efforts. Based on the BRAF-KD/1 structure and the similarity between inhibitors 1 and 2 (Figure 2A), we hypothesized that the naphthol heterocycle of 2 would occupy the same space as the quinolol of 1 with the thiotriazole and the thienylsulfonamide of 2 occupying the positions of the furan/aminopyridine and chloride groups of 1, respectively. Based on this hypothesis, we resynthesized 2 and prepared a total of 38 compounds varying the nature of both the nitrogen-containing heterocycle corresponding to the thiotriazole ring and the sulfonamide substitution in 2. The general scheme for the synthesis of these compounds is shown in Figure 2B. We then screened each of these analogues against BRAFV600E and BRAFWT (at protein and ATP concentrations of 28 nM and 100 uM, respectively) using a dose response curve to calculate IC50 values.
As can be seen in Figure 5 and Table 2, resynthesized 2 shows IC50 values for BRAFV600E and BRAFWT of 0.17 μM and 0.26 μM, respectively, close to the values of 0.11 μM and 0.32 μM obtained for the screen compound. To further confirm our modeling of 2 into the ATP binding site of BRAFWT based on the BRAFWT/1 structure we prepared the oxidized form of 2 (compound 24). Based on this modeling, we hypothesized that the OH group of the naphthol points into solvent so inhibitor potency would therefore be insensitive to its oxidation state. Our data showed that 2 and 24 show comparable potencies for BRAFWT and BRAFV600E (Table 2), This is consistent with our hypothesis. Nonetheless, it is possible that the assay conditions may regenerate the oxidized form, therefore leaving open the possibility that BRAFWT/BRAFV600E actually binds to the oxidized form of the naphthol analogues prepared in this study.
An SAR of the other naphthol analogues of 2 reveals that, as expected, other substitutions on the naphthol ring system have significant effects on inhibitor potency for BRAFV600E (Table 2 and Supplemental Table 1). Specifically, replacement of the thiotriazole present in 2 with a hydrogen atom reduced potency by more than 30-fold (compare 2 with 25) whereas replacement of the thiotriazole with a halogen only reduced binding by about 3-fold (compare 2 with 26 and 27). In contrast, several different heterocycles at this position can be accommodated (28 – 31) while another compound is more poorly accommodated (32). The structure of the BRAFWT/1 complex suggests that the thiotriazole of the modeled 2 would be pointing out towards the solvent of the enzyme where the protein substrate would bind suggesting that these substitutions may have differential effects on substrate binding. Substitution of the sulfonamide that presumably occupies the RAF-specific pocket also has significant effects on inhibitor potency. Specifically, sulfonamides that contain a functionalized aryl group (for example, 33, 34, 40 and 41) are significantly more potent than sulfonamides containing long aliphatic groups (for example, 35 and 36). In addition, highly substituted aryls in the sulfonamide moiety significantly decrease inhibitor potency by 3 to 20-fold (for example, 37 – 39). We hypothesize that the highly substituted aryls or long aliphatic groups are too bulky or long, respectively, to be accommodated by the BRAFWT/BRAFV600E specificity pocket. The most potent BRAFV600E inhibitors identified were 28 (IC50 = 0.12 μM), 44 (IC50 = 0.14 μM) and 40 (IC50 = 0.08 μM) (Figure 5 and Table 3). Each of these inhibitors exhibited about a 2-fold selectivity for BRAFV600E over BRAFWT.
In order to survey the selectivity of the most potent BRAFV600E inhibitor, 40, we profiled the compound against 7 other randomly chosen kinases representing each of the 6 major groups of serine-threonine kinases; PAK1 (STE group), PIM1 (CAMK group), GSK3 (CMGC group), p70S6K (AGC group), PI3Kα and PI3Kγ (CK1 group) and the BRAF related CRAF kinase (TKL group), at an inhibitor concentration of 1 μM. This analysis revealed that 40 showed significant selectivity for BRAFV600E and BRAFWT over the other kinases (Figure 6A). While BRAFV600E and BRAFWT both showed less than 10% remaining activity, CRAF showed about 60% remaining activity, while the other kinases showed 75% or greater remaining activity. Taken together, these data demonstrates that 40 has significant selectivity for BRAFV600E and BRAFWT over representative kinases from the other major groups of serine-threonine kinases.
We further investigated the effects of 40 in the context of human-derived skin and melanoma cells. A proliferation assay indicated that a mutant BRAF melanoma cell line (451Lu, harboring a BRAFV600E mutation) is more sensitive to the compound than its normal melanocyte and fibroblast counterparts (Figure 6B). Western blot analyses further indicated that MAPK pathway activity is decreased, as evidenced by reduced phospho-MEK levels in the presence of the inhibitor, compared to the vehicle control treated cells (Figure 6C).
In this study, we report on developing an ELISA based high-throughput assay that is suitable for screening a large library of small molecule compounds for inhibition of BRAF kinase activity. We employed this assay to screen a combined diverse library of over 30,000 organic compounds for BRAFV600E inhibition. Of the inhibitors that were identified, we further pursued a family of related quinolol and naphthol compounds that appeared to represent a novel family of BRAFV600E inhibitors. We then determined the X-ray crystal structure of one of these inhibitors bound to the BRAF kinase domain revealing that the compounds bind in the ATP binding cleft of the kinase in its active conformation. This structure was used as a scaffold to initiate a medicinal chemistry campaign that resulted in the preparation of a family of naphthol compounds that selectively inhibit BRAFV600E over BRAFWT in vitro with IC50 values in the 80–200 nM range in the presence of 100 uM ATP. We also demonstrated that one of these compounds, 40, has significant selectivity for BRAFV600E and BRAFWT over other kinases and inhibits MAPK signaling in melanoma cells.
The medicinal chemistry campaign revealed that substitutions on the naphthol ring system had significant effects on inhibitor potency for BRAFV600E. Most significantly, substitutions in the naphthol ring system corresponding to the thienylsulfonamide present in 2 (the most potent compound obtained from the primary screen) had the most dramatic effects on inhibitor potency (Table 2 and Supplemental Table 1). Modeling of 2 onto the X-ray structure of the BRAF-KD/1 complex (where the thienylsulfonamide of 2 overlays with the chloride group of 1) suggests that this sensitive position of the inhibitor points into a pocket that is unique to BRAF kinase (Figure 4C). Indeed, the superposition of the BRAF complexes with 1 and PLX4720 confirms this as it reveals that the chloride group of 1 points towards the sulfonamide group of PLX4720 that sits in the BRAF specificity pocket (Figure 7). In light of this, it is not a surprise that different sulfonamides at this position on the backbone of 2 have dramatic effects on inhibitor potency, with sulfonamides that contain a functionalized aryl group yielding more potent inhibitors than sulfonamides containing long aliphatic groups. Clearly, the preparation of other sulfonamide analogues of 2 might yield more potent inhibitors and hybrid molecules containing the 2 backbone with the sulfonamide derivative of PLX4720 could provide particularly potent and selective BRAFV600E inhibitors.
Initial in vitro and cell-based assays indicate that our current approach yielded a BRAF/BRAFV600E selective inhibitor (40) that reduces melanoma cell proliferation preferentially over primary fibroblasts and melanocytes. However, the downstream effector of BRAF (MEK) appears to be inhibited regardless of the cells’ transformation or BRAF status. It is therefore possible that these inhibitors might be hitting other pathways in cells that are important for melanoma cell proliferation and a broader kinase profiling and cell based studies would have to be carried out to address this possibility. In addition, our cell based studies reveal that MEK appears to be activated at higher inhibitor concentrations (Figure 6C), suggesting that high inhibitor concentration may lead to the transactivation of BRAF/CRAF dimers as has been noted elsewhere 20, 21. Given this observation, the further development of these inhibitors could take two paths. In one scenario, these inhibitors could be further optimizated towards increased BRAFV600E potency and specificity to develop an alternative BRAFV600E inhibitor that might have more favorable properties than other BRAFV600E selective inhibitors. Greater BRAFV600E potency and specificity could be achieved by further filling the BRAF/BRAFV600E specificity pocket (Figure 7). Alternatively, these inhibitors could be optimized to also inhibit CRAF, as our kinase profiling studies do show some inhibition of 40 against CRAF. Such compounds might be particularly effective against reactivation of MAPK pathway activity through the transactivation of BRAF/CRAF dimers. Taken together, this study has resulted in the identification of a new lead series of BRAF inhibitors with the potential for further preclinical development for therapeutic use in melanoma.
The human BRAF kinase domain and the V600E containing mutant (BRAFWT and BRAFV600E, residues 433–726) with an N-terminal 6X-His tag (MDRGSH6GS) to facilitate protein purification (in the presence of full-length mouse p50cdc37 to faciliate the proper folding of the BRAF kinase) was expressed in Sf9 cells and purified to homogeneity as previously described 19. The protein was stored at 1.5 mg/mL at 4 °C until use. GST-MEK-His protein was overexpressed at 37 °C in Escherichia coli BL21 (Gold) cells as previously described 19. The protein was stored at 10 mg/mL at −80°C until use.
Recombinantly expressed GST MEK-His, diluted in TTBS buffer [20 mM Tris (pH 7.5), 150 mM NaCl, and 0.05% Tween 20] to 50 μg/mL in a volume of 100 μL, was bound to the wells of a 96-well glutathione-coated plate (Pierce Biotechnology). One microliter of compound (as racemic mixtures) with 2 serial dilutions in a 100% DMSO stock solution was added to a mixture of 50 μL of a buffer containing 50mMHEPES (pH7.0) with 0.7 pmol of BRAFV600E kinase. This mixture was incubated at room temperature for 1 h before it was added to the GST-MEK-His-bound wells of the 96-well plate. An additional 50 μL of phosphorylation buffer [50 mM HEPES (pH 7.0), 200 mM NaCl, 10 mM MgCl2, and 200 μM ATP] was added to the well mixture to start the kinase reaction at 37 °C for 30 min. with intermittent shaking. The kinase reaction was stopped by extensive washing with TTBS buffer, and a 1:5000 dilution of anti-phospho-MEK1 (Ser218/222)/MEK2 (Ser222/226) monoclonal antibody (Millipore) in TTBS buffer was subsequently added to the wells and incubated for 1 h with shaking. Goat anti-rabbit IgG (H+L)-HRP conjugate (Bio-Rad Laboratories) in a 1:5000 dilution was added to the wells for incubation at room temperature with shaking. Finally, the SuperSignal ELISA Pico chemiluminescent substrate (Pierce Biotechnology) was added to the wells. The luminescence signal was recorded with a luminescence filter using a Wallac 1420 luminometer (PerkinElmer).
High throughput inhibitor screening was performed at the Broad Institute of Harvard and MIT screening center. Assays were conducted in glutathione coated 384-well plates and followed the procedures essentially as described above but using a 50 μL reaction volume instead of a 100 μL reaction volume to fit 384-well plate format. Specifically, GST-MEK protein diluted in TTBS to 50 μg/mL was dispensed into the wells of the glutathione coated 384-well plate to a final volume of 50 μL/well using a Matrix Wellmate Dispanser with a microplate stacker (Thermo Scientific). Each plate was agitated using an orbital shaker at 2,500 rpm for 1 min and incubated at room temperature for 1 hour. Plates were aspirated and washed once using a wash program with vigorous agitation using an automated microplate washer (Biotek). 3.5 pmoles of BRAFV600E kinase domain diluted in 25 μL of 50 mM HEPES pH 7.5 buffer was added into each well of the plate using the Matrix Wellmate Dispanser, and 25 nL of individual compound (10 mM in 100% DMSO) was transferred into the solution using a Cybi-Well pin-transfer station (Cybio). Plates were agitated using an orbital shaker for 1 min and incubated at room temperature for 1 hour. 25 μL of phosphorylation buffer was then added into the wells to start the kinase reaction (controls were tested to ensure the precise timing for the start of the kinase reaction). Kinase reactions were conducted at room temperature for 30 min and stopped by washing using the microplate washer. A 1:5,000 dilution of Anti-phospho-MEK1 (Ser218/222)/MEK2 (Ser222/226) monoclonal antibody (Millipore) in TTBS buffer was subsequently dispensed into the wells to a final volume of 50 μL and incubated for 1 hour with shaking. Goat anti-rabbit IgG (H+L)-HRP conjugate (BioRad Laboratories) in a 1:5,000 dilution was then dispensed into the wells to a final volume of 50 μL to incubate at room temperature with agitation. Finally, 50 μL of the SuperSignal ELISA Pico chemiluminescent substrate (Pierce Biotechnology) was dispensed into the wells to generate the chemiluminescence signal, which was detected using a 700 nm luminescence filter by an Envision chemiluminescence detector (PerkinElmer). A total of 31,976 compounds were screened in duplicate including libraries of a diversity oriented synthesis (DOS), commercially available drug-like compounds, bioactive compounds, natural products, compounds collected from academic organic synthesis laboratories and a ChemBridge Kinase inhibitor biased library. Compounds were ranked based on a composite Z-score of both duplicates and the top 100 compounds were cherry-picked from the compound plates at the Broad Institute and they were re-analyzed by the same assay to confirm their inhibitory activities. According to the results, the top 23 compounds that were deemed to have drug-like properties were confirmed by reordering the compounds from their source vendors and confirming their inhibitor activities.
For IC50 calculations of the related quinolol and naphthol inhibitors, the same assay described above was used at different inhibitor concentrations to generate a sigmoidal does response curve using BRAFV600E or BRAFWT protein. All dose response measurements were carried out in duplicate or triplicate and IC50 values were derived from fitting the data to a sigmoidal dose response curve with a four-parameter logistic model using GraphPad Prism.
Compound 1 dissolved in 100% DMSO was added to 1.5 mg/mL BRAFWT solution in 4-fold molar excess and this mixture was incubated at room temperature for 1 hour before the precipitate was removed by high speed centrifugation. The supernatant was used in the crystallization trials. Crystals were obtained by mixing 1.35 μL of the BRAFWT/1 complex with 1.35 μL of crystallization reservoir solution containing 200 mM magnesium acetate tetrahydrate, 100 mM sodium cacodylate trihydrate pH 6.5, 20% polyethylene glycol (PEG) 8,000) (Hampton Research) supplemented with 0.3 μL of 100 mM nicotinamide adenine dinucleotide (Hampton Research) as an additive by the microbatch method underneath 6 mL of light mineral oil (Sigma) in a 12 × 6 microbatch plate (Hampton Research). Crystals reached a maximum size of 30 μm × 30 μm × 200 μm after about 1 week. Crystals were washed and harvested in cryoprotecting harvest solution containing 200 mM magnesium acetate tetrahydrate, 100 mM sodium cacodylate trihydrate pH 6.5, 25% PEG 8,000 and 15% glycerol, and flash frozen in liquid propane. Data was collected at beamline GM/CA-CAT 23ID-B at the Advanced Photon Synchrotron Source (Argonne National Laboratory). We discovered that although the crystals showed diffraction to 3.5 Å using a typical X-ray beam (300 μm), they suffered very badly from radiation decay. We were not able to collect more than 5 frames of diffraction data using this size X-ray beam. In contrast, we found that by using an X-ray beam of 10 by 10 μm (minibeam setup at APS beamline 23ID-B), we were able to collect about 20–30 images per crystal position and both the resolution and mosaicity were significantly improved. Also importantly, the minibeam setup allowed us to move the beam center to new, unexposed parts of the crystal, allowing us to collect a complete data set from a single crystal. Diffraction data was indexed, integrated and scaled using the HKL2000 package (HKL Research) and was further processed using the CCP4 program suite 22.
The structure was determined by molecular replacement using the program Molrep 23 from the CCP4 suite using a previously determined BRAF structure (PDB 1UWH) as a search model. The spacegroup was determined to be P41212 and each asymmetric unit contained two molecules. Electron density corresponding to the organic BRAF inhibitor 1 was well resolved in both molecules of the asymmetric unit and the ligand model was placed into the electron density from the calculated Fo − Fc map and adjusted in Coot 24. Parameter and topology files for 1 used in the refinements were generated from HIC-UP XDICT server (http://alpha2.bmc.uu.se/hicup/xdict.html). This was followed by additional refinement using CNS 25 and the final model was checked for errors using an CNS composite omit map for the protein model and a simulated annealing omit map for inhibitor. The final model was refined with excellent geometry and refinement statistics (Table 1).
Solvents used for extraction and purification were HPLC grade from Fisher. Unless otherwise indicated, all reactions were run under an inert atmosphere of argon. Anhydrous tetrahydrofuran, ethyl ether, and toluene were obtained via passage through an activated alumina column. Merck pre-coated silica gel plates (250 mm, 60 F254) were used for analytical TLC. Spots were visualized using 254 nm ultraviolet light, with either anisaldehyde or potassium permanganate stains as visualizing agents. Chromatographic purifications were performed on Sorbent Technologies silica gel (particle size 32–63 microns). 1H and 13C NMR spectra were recorded at 500 MHz and 125 MHz, or 360 MHz and 90 MHz, respectively, in CDCl3, DMSO-d6, or CD3OD on a Bruker AM-500, a DRX-500, or a DMX-360 spectrometer. Chemical shifts are reported relative to internal chloroform (d 7.26 for 1H, d 77.0 for 13C), DMSO-d6 (d 2.50 for 1H, d 39.5 for 13C), or CD3OD (d 3.31 for 1H, d 49.0 for 13C). Infrared spectra were recorded on a NaCl plate using a Perkin-Elmer 1600 series Fourier transform spectrometer. High resolution mass spectra were obtained by Dr. Rakesh Kohli at the University of Pennsylvania Mass Spectrometry Service Center on an Autospec high resolution double-focusing electrospray ionization/chemical ionization spectrometer with either DEC 11/73 or OPUS software data system. Melting points were obtained on a Thomas Hoover capillary melting point apparatus and are uncorrected. Based on the analytical methods described above, all compounds were judged to have purities of ≥ 95 %.
General procedure for the synthesis of naphthol derivatives was adapted from WO 2010/005534 26. Unless otherwise specified, analogues were synthesized using the general procedure.
2: Spectroscopic data was identical to that reported previously (WO Patent 2010/005534, January 14, 2010) 26.
24: Spectroscopic data was identical to that reported previously (WO Patent 2010/005534, January 14, 2010) 26.
25: To a mixture of 4-amino-1-naphthol•HCl (0.196 g, 1.00 mmol), pyridine (0.17 mL, 2.1 mmol), and DMAP (0.005 g, 0.04 mmol) in CH2Cl2 was added 2-thiopenesulfonyl chloride (0.194 g, 1.06 mmol). The mixture was heated at reflux for 18 h. The mixture was allowed to cool to 25 °C, and solvent was removed in vacuo. The resulting residue was diluted with EtOAc (20 mL), and the solution was washed with water (2×10mL), washed with brine, dried over Na2SO4, filtered, and concentrated to afford a brown residue. The residue was purified by flash chromatography (5:95 MeOH:CH2Cl2) to afford the product as a tan solid (0.240 g, 79%): 1H NMR (500 MHz, DMSO-d6) δ 10.32 (s, 1H), 9.99 (s, 1H), 8.10-8.08 (m, 1H), 7.84 (dd, J = 5.0, 1.4 Hz, 2H), 7.43-7.37 (m, 2H), 7.35 (dd, J = 3.7, 1.4 Hz, 1H), 7.06 (dd, J = 5.0, 3.7 Hz, 1H), 6.97 (d, J = 8.1 Hz, 1H), 6.76 (d, J = 8.1 Hz, 1H); 13C NMR (125 MHz, DMSO-d6) δ 153.3, 141.3, 133.5, 132.6, 132.2, 128.1, 126.8, 126.6, 125.5, 125.4, 123.6, 123.4, 122.7, 107.9; IR (thin film) 3268, 3115, 1699, 1584, 1317, 1153, 1016, 774 cm−1; HRMS (ESI) m/z calcd for C14H11NO3NaS2 (M + Na)+ 328.0078, found 328.0078.
26: To a solution of 2-chloro-1,4-naphthoquinone27 (0.119 g, 0.50 mmol) and 2-thiophene sulfonamide (0.098 g, 0.60 mmol) cooled to 0 °C in THF (1.8 mL) were added titanium (IV) chloride tetrahydrofuran complex (0.200 g, 0.60 mmol), followed by the addition of triethylamine (0.18 mL, 1.3 mmol). The mixture was allowed to warm to 25 °C and was heated to 60 °C using a microwave (80 watts) for 1 h. The mixture was cooled to 25 °C and added to EtOAc (20 mL). The mixture was filtered though a pad of Celite with EtOAc. The filtrate was concentrated to dryness in vacuo. The resultant residue was suspended in CH2Cl2 (10 mL) and filtered though a pad of Celite. The filtrate was concentrated in vacuo to dryness, and the resulting residue was suspended in EtOAc (10 mL). To the suspension was added Na2S2O4 (0.436 g, 2.50 mmol) followed by H2O (5 mL). The mixture was stirred at 25 °C for 1 h. The layers were separated, and aqueous layer was extracted with EtOAc (3 × 10 mL). The combined organic layers were washed with brine (25 mL), dried over Na2SO4, filtered, and concentrated to afford a brown residue. The residue was purified by flash chromatography (40:60 EtOAc:hexanes) to afford the product as a tan solid (0.081 g, 47%): 1H NMR (500 MHz, DMSO-d6) δ 10.27 (s, 2H), 8.17 (d, J = 8.3 Hz, 1H), 7.87 (d, J = 8.4 Hz, 1H), 7.83 (d, J = 4.4 Hz, 1H), 7.50 (d, J = 7.6 Hz, 1H), 7.46-7.41 (m, 2H), 7.07 (s, 1H), 7.05 (s, 1H);); 13C NMR (125 MHz, DMSO-d6) δ 148.4, 140.5, 133.9, 133.1, 130.7, 128.3, 127.3, 127.0, 126.54, 126.47, 125.0, 123.8, 122.8, 113.7; IR (thin film) 3276, 1581, 1458, 1401, 1315, 1153, 849, 759 cm−1; HRMS (ESI) m/z calcd for C14H9NO3S2Cl (M − H)− 337.9712, found 337.9714.
27: To a solution of 2-bromo-1,4-naphthoquinone 28 (0.119 g, 0.50 mmol) and 2-thiophene sulfonamide (0.098 g, 0.60 mmol) cooled to 0 °C in THF (1.8 mL) were added titanium (IV) chloride tetrahydrofuran complex (0.200 g, 0.60 mmol), followed by the addition of triethylamine (0.18 mL, 1.3 mmol). The mixture was allowed to warm to 25 °C and was heated to 60 °C using a microwave (80 watts) for 1 h. The mixture was cooled to 25 °C and added to EtOAc (20 mL). The mixture was filtered though a pad of Celite with EtOAc. The filtrate was concentrated to dryness in vacuo. The resultant residue was suspended in CH2Cl2 (10 mL) and filtered though a pad of Celite. The filtrate was concentrated in vacuo to dryness, and the resulting residue was suspended in EtOAc (10 mL). To the suspension was added Na2S2O4 (0.436 g, 2.50 mmol) followed by H2O (5 mL). The mixture was stirred at 25 °C for 1 h. The layers were separated, and aqueous layer was extracted with EtOAc (3 × 10 mL). The combined organic layers were washed with brine (25 mL), dried over Na2SO4, filtered, and concentrated to afford a brown residue. The residue was purified by flash chromatography (40:60 EtOAc:hexanes) to afford the product as a brown solid (0.036 g, 19%): 1H NMR (500 MHz, DMSO-d6) δ 10.26 (s, 1H), 10.24 (s, 1H), 8.18 (d, J = 8.4 Hz, 1H), 7.88-7.86 (m, 2H), 7.51 (ddd, J = 8.3, 7.0, 1.2 Hz, 1H), 7.45 (ddd, J = 8.4, 6.9, 1.3 Hz, 1H), 7.40 (dd, J = 3.7, 1.3 Hz, 1H), 7.08 (dd, J = 5.0, 3.7 Hz, 1H), 7.06 (s, 1H); 13C NMR (125 MHz, DMSO-d6) δ 147.8, 140.1, 133.4, 132.5, 130.1, 127.7, 126.6, 126.3, 126.0, 125.9, 124.5, 123.3, 122.2, 113.2; IR (thin film) 3479, 3098, 1582, 1459, 1415, 1315, 1154, 759 cm−1; HRMS (ESI) m/z calcd for C14H9NO3S2Cl (M – HBr + Cl)− 337.9712, found 337.9707.
28: 1H NMR (500 MHz, DMSO-d6) δ 10.09 (s, 1H), 8.20 (d, J = 8.0 Hz, 1H), 7.88-7.82 (m, 2H), 7.53-7.47 (m, 1H), 7.47-7.41 (m, 1H), 7.31 (dd, J = 1.5, 3.5 Hz, 1H), 7.21 (s, 2H), 7.05 (dd, J = 3.5, 5.0 Hz, 1H), 6.83 (s, 1H); 13C NMR (125 MHz, DMSO-d6) δ 152.2, 145.1, 137.4, 131.4, 131.0, 129.8, 129.2, 127.2, 126.1, 125.7, 123.9, 123.4, 123.1, 122.6; IR (KBr disk) 3262, 3105, 1709, 1567, 1506, 1451, 1403, 1377, 1325, 1157, 1016, 911, 854, 766, 723, 669 cm−1; HRMS (ESI) m/z calcd for C17H14N3O3S3 (M + H)+ 404.0197, found 404.0204.
29: 1H NMR (500 MHz, CDCl3) δ 10.85 (s, 1H), 8.44 (d, J = 5.0 Hz, 1H), 8.39 (d, J = 7.5 Hz, 1H), 7.76 (d, J = 7.5 Hz, 1H), 7.61 (t, J = 7.5 Hz, 1H), 7.53-7.43 (m, 3H), 7.41 (s, 1H), 7.36 (d, J = 4.0 Hz, 1H), 7.24 (s, 1H), 7.16 (dd, J = 4.0, 7.5 Hz, 1H), 6.89 (dd, J = 3.5, 5.0 Hz, 1H), 6.61 (br s, 1H); 13C NMR (125 MHz, CDCl3) δ 159.8, 156.4, 149.1, 139.8, 138.1, 133.2, 132.6, 131.1, 128.5, 127.5, 126.4, 126.3, 124.0, 123.5, 123.1, 121.9, 121.6, 110.1; IR (KBr disk) 3235, 3062, 2920, 2848, 1564, 1451, 1414, 1374, 1322, 1153, 1070, 1015, 762, 719 cm−1; HRMS (ESI) m/z calcd for C19H14N2O3S3 (M + H)+ 415.0245, found 415.0257.
30: Spectroscopic data was identical to that reported previously (WO Patent 2010/005534, January 14, 2010) 26.
31: 1H NMR (500 MHz, CDCl3) δ 8.31 (d, J = 8.0 Hz, 1H), 7.98 (d, J = 8.0 Hz, 1H), 7.62-7.53 (m, 2H), 7.42 (dd, J = 1.0, 5.0 Hz, 1H), 7.35 (dd, J = 1.0, 3.5 Hz, 1H), 7.30-7.16 (m, 5H), 7.10-7.06 (m, 2H), 6.83 (dd, J = 4.0, 5.0 Hz, 1H), 6.62 (br s, 1H); 13C NMR (125 MHz, DMSO-d6) δ 154.9, 139.4, 135.5, 133.3, 133.0, 132.6, 131.5, 129.4, 128.8, 127.6, 127.4, 126.7, 124.3, 124.0, 123.8, 122.6, 108.9; IR (KBr disk) 3391, 3272, 2920, 2851, 1568, 1435, 1373, 1346, 1311, 1266, 1220, 1156, 1070, 907, 720 cm−1; HRMS (ESI) m/z calcd for C20H15NNaO3S3 (M + H)+ 436.0112, found 436.0120.
32: 1H NMR (500 MHz, DMSO-d6) δ 10.50-9.50 (br, 2H, contains two s), 8.16 (d, J = 8.5 Hz, 1H), 8.15-7.75 (m, 3H, contains one br s; one d, J = 8.5 Hz; one dd, J = 1.5, 5.0 Hz), 7.54-7.47 (m, 1H), 7.47-7.40 (m, 1H), 7.26 (dd, J = 1.5, 3.5 Hz, 1H), 7.02 (dd, J = 3.5, 5.0 Hz, 1H), 6.68 (s, 1H); 13C NMR (125 MHz, DMSO-d6) δ 140.1, 133.0, 132.1, 130.7, 127.5, 126.4, 126.0, 125.5, 124.6, 123.5, 122.1; IR (KBr disk) 3101, 2991, 2917, 1568, 1506, 1452, 1402, 1376, 1323, 1263, 1156, 1072, 1020, 912, 854, 764, 723, 670 cm−1; HRMS (ESI) m/z calcd for C16H13N4O3S3 (M + H)+ 405.0150, found 405.0166.
33: 1H NMR (500 MHz, DMSO-d6) δ 10.13 (s, 1H), 9.73 (s, 1H), 8.54 (s, 1H), 8.19-8.17 (m, 1H), 8.02-7.99 (m, 2H), 7.52-7.47 (m, 2H), 7.44-7.42 (m, 1H), 7.40 (d, J = 0.3 Hz, 1H), 6.89 (s, 1H), 4.11 (t, J = 8.5 Hz, 2H), 3.06 (t, J = 8.4 Hz, 2H), 2.18 (s, 3H); 13C NMR (125 MHz, DMSO-d6) δ 169.9, 135.1, 134.7, 134.1, 133.1, 127.3, 127.2, 126.5, 126.3, 126.2, 126.0, 124.1, 123.9, 123.8, 122.8, 115.4, 113.6, 111.3, 49.0, 27.2, 24.5; IR (thin film) 3121, 2924, 1708, 1660, 1481, 1397, 1149, 1072 cm−1; HRMS (ESI) m/z calcd for C22H19N5O4NaS2 (M + Na)+ 504.0776, found 504.0788.
34: 1H NMR (500 MHz, DMSO-d6) δ 14.00 (br s, 1H), 10.80-9.60 (br s, 2H), 8.46 (br s, 1H), 8.18 (d, J = 8.0 Hz, 1H), 7.93-7.73 (m, 5H), 7.55-7.37 (m, 2H), 6.88 (br s, 1H); 13C NMR (125 MHz, DMSO-d6) δ 152.2, 145.1, 143.5, 132.7, 132.4, 132.2, 131.3, 129.1, 127.7, 126.9, 126.2, 126.1, 125.9, 125.6, 124.5, 123.8, 123.2, 122.6, 122.3; 19F NMR (470 MHz, DMSO-d6) δ −61.6; IR (KBr disk) 3102, 2989, 2928, 1730, 1568, 1504, 1453, 1403, 1376, 1322, 1267, 1168, 1062, 762, 711 cm−1; HRMS (ESI) m/z calcd for C19H14F3N4O3S2 (M + H)+ 467.0459, found: 467.0471.
35: 1H NMR (500 MHz, DMSO-d6) δ 9.42 (s, 1H), 8.49 (s, 1H), 8.23 (d, J = 8.1 Hz, 1H), 8.17 (d, J = 8.5 Hz, 1H), 7.62-7.55 (m, 2H), 7.29 (s, 1H), 3.00 (dd, J = 7.9, 7.7 Hz, 2H), 1.70-1.64 (m, 2H), 1.30-1.17 (m, 14H), 0.85 (t, J = 6.9 Hz, 3H); 13C NMR (125 MHz, DMSO-d6) δ 151.9, 133.1, 132.1, 131.6, 130.2, 128.7 127.1, 126.0, 125.0, 124.5, 123.8, 122.6, 50.7, 31.3, 28.87, 28.68, 28.48, 27.5, 23.1, 22.1, 14.0; IR (thin film) 2923, 2853, 1567, 1454, 1279, 1141, 1076, 767 cm−1; HRMS (ESI) m/z calcd for C22H30N4O3NaS2 (M + Na)+ 485.1657, found 485.1656.
36: 1H NMR (500 MHz, DMSO-d6) δ 10.23 (s, 1H), 9.44 (s, 1H), 8.62 (s, 1H), 8.24 (d, J = 8.3 Hz, 1H), 8.18 (d, J = 8.1 Hz, 1H), 7.62-7.55 (m, 2H), 7.31 (s, 1H), 3.01 (t, J = 7.8 Hz, 2H), 1.71-1.65 (m, 2H), 1.30-1.19 (m, 10H), 0.84 (t, J = 7.0 Hz, 3H); 13C NMR (125 MHz, DMSO-d6) δ 145.1, 140.5, 134.4, 131.6, 127.2, 127.1, 126.1, 125.8, 123.9, 122.6, 50.7, 31.2, 28.5, 28.4, 27.5, 23.2, 22.1, 14.0; IR (thin film) 3253, 2925, 2855, 1568, 1454, 1279, 1141, 766 cm−1; HRMS (ESI) m/z calcd for C20H26N4O3NaS2 (M + Na)+ 457.1344, found 457.1325.
37: 1H NMR (500 MHz, CD3OD) δ 8.25 (dt, J = 4.4, 2.4 Hz, 1H), 8.19 (s, 1H), 8.06 (dt, J = 4.3, 2.4 Hz, 1H), 7.42-7.41 (m, 2H), 7.38 (s, 1H), 7.18-7.08 (m, 10H), 6.90 (dt, J = 8.8, 4.4 Hz, 1H), 6.63 (t, J = 9.4 Hz, 1H), 4.02 (s, 4H); 13C NMR (125 MHz, CD3OD) δ 156.2, 155.7, 155.3, 154.2, 153.3, 138.6, 137.1, 133.5, 130.8, 129.3, 128.9, 128.6, 128.5, 128.2, 127.2, 127.1, 125.2, 124.3, 123.8, 119.1, 112.8, 112.6, 68.8, 57.0, 26.4; IR (thin film) 3087, 2924, 1705, 1570, 1485, 1364, 1171, 745 cm−1; HRMS (ESI) m/z calcd for C32H25N5O3F2NaS2 (M + Na)+ 652.1265, found 652.1271.
38: 1H NMR (500 MHz, DMSO-d6) δ 10.98 (s, 1H), 10.38 (s, 1H), 8.59 (s, 1H), 8.24-8.22 (m, 1H), 7.99 (s, 1H), 7.58 (s, 2H), 6.97 (s, 1H); 13C NMR (125 MHz, DMSO-d6) δ 157.1, 152.4, 145.5, 143.2, 138.8, 136.8, 131.8, 128.8, 127.7, 126.7, 126.1, 123.5, 123.0, 115.6, 112.1; IR (thin film) 3129, 2928, 1704, 1520, 1502, 1176, 1101, 992 cm−1; HRMS (ESI) m/z calcd for C18H10N4O3F5S2 (M + H)+ 489.0115, found 489.0110.
39: 1H NMR (500 MHz, DMSO-d6) δ 10.76 (s, 1H), 10.31 (s, 1H), 8.57 (s, 1H), 8.23-8.21 (m, 1H), 7.93 (dd, J = 6.6, 2.9 Hz, 1H), 7.69-7.65 (m, 1H), 7.53 (dd, J = 6.4, 3.2 Hz, 2H), 7.32 (t, J = 9.0 Hz, 1H), 7.16 (s, 1H), 2.90-2.82 (m, 4H); 13C NMR (125 MHz, DMSO-d6) δ 176.2, 159.6, 157.5, 155.4, 153.3, 145.5, 135.6, 131.8, 129.8, 127.7, 126.4, 126.0, 123.4, 123.0, 119.0, 118.4, 114.0, 113.7, 29.1; IR (thin film) 2920, 2848, 1719, 1592, 1562, 1492, 1298, 1172 cm−1; HRMS (ESI) m/z calcd for C22H15N5O5NaS2F2 (M + Na)+ 554.0380, found 554.0386.
40: To a solution of 2-chloro-1,4-naphthoquinone27 (0.096 g, 0.50 mmol) and biphenyl-4-sulfonamide (0.140 g, 0.60 mmol) cooled to 0 °C in THF (1.8 mL) was added titanium (IV) chloride tetrahydrofuran complex (0.200 g, 0.60 mmol), followed by the addition of triethylamine (0.18 mL, 1.3 mmol). The mixture was allowed to warm to 25 °C and was heated to 60 °C using a microwave (80 watts) for 1 h. The mixture was cooled to 25 °C and added to EtOAc (20 mL). The mixture was filtered though a pad of Celite with EtOAc. The filtrate was concentrated to dryness in vacuo. The resultant residue was suspended in CH2Cl2 (10 mL) and filtered though a pad of Celite. The filtrate was concentrated to dryness and then dissolved in THF (10 mL). To the solution was added 1H-1,2,4-triazole-3-thiol (0.051 g, 0.50 mmol), and the mixture was stirred at 25 °C for 4 h. The mixture was concentrated in vacuo to dryness, and the resultant residue was suspended in EtOAc (10 mL). To the suspension was added Na2S2O4 (0.436 g, 2.50 mmol) followed by H2O (5 mL). The mixture was stirred at 25 °C for 1 h. The layers were separated, and aqueous layer was extracted with EtOAc (3 × 10 mL). The combined organic layers were washed with brine (25 mL), dried over Na2SO4, filtered, and concentrated to afford a brown residue. The residue was purified by flash chromatography (3:97 MeOH:CH2Cl2) to afford the product as a colorless solid (0.111 g, 47%): 1H NMR (500 MHz, DMSO-d6) δ 10.22 (s, 0.5H), 9.98 (s, 1H), 8.54 (s, 0.5H), 8.19 (d, J = 7.9 Hz, 1H), 7.97 (d, J = 8.0 Hz, 1H), 7.74 (d, J = 8.5 Hz, 2H), 7.68 – 7.66 (m, 5H), 7.51-7.41 (m, 6H), 6.96 (s, 1H); 13C NMR (125 MHz, DMSO-d6) δ 144.1, 138.6, 138.4, 129.2, 128.5, 127.5, 127.2, 127.1, 127.0, 126.0, 125.7, 123.5, 122.6; IR (thin film) 3249, 2923, 1704, 1593, 1452, 1320, 1160, 763 cm−1; HRMS (ESI) m/z calcd for C24H18N4O3NaS2 (M + Na)+ 497.0718, found 497.0711.
41: 1H NMR (360 MHz, DMSO-d6) δ 14.11 (br s, 1H), 10.60-9.70 (br s, 2H), 8.32 (br s, 1H), 8.21 (s, 1H), 8.15 (d, J = 8.0 Hz, 1H), 8.15-7.75 (m, 4H), 7.80-7.55 (m, 3H), 7.55-7.30 (m, 2H), 6.92 (br s, 1H); 13C NMR (125 MHz, DMSO-d6) δ 151.8, 145.0, 136.9, 134.1, 131.5, 131.4, 129.2, 129.1, 128.7, 128.6, 127.8, 127.7, 127.5, 126.8, 125.9, 125.6, 124.4, 123.5, 122.4, 122.3; IR (KBr disk) 3259, 3053, 2851, 1564, 1508, 1468, 1450, 1347, 1318, 1285, 1156, 1026, 945, 766, 666 cm−1; HRMS (ESI) m/z calcd for C22H17N4O3S2 (M + H)+ 449.0742, found 449.0753.
The kinase assays for PAK1, PIM1, GS3Kα, PI3Kα and PI3Kγ were carried out as described previously 29 and CRAF was assayed as described above for BRAF using between 20–30 nM enzyme and 100 μM ATP.
The p70S6K kinase assay was performed using labeled γ-P32 ATP and the incorporation of labeled phosphate onto substrate (S6 kinase/Rsk2 peptide) was monitored. Various concentrations of inhibitor were incubated at room temperature in 20 mM MOPS, 30 mM MgCl2, 0.8 μg/μL BSA, 5% DMSO (resulting from the inhibitor stock solution), pH 7.0, in the presence of substrate (S6 kinase/Rsk2 Substrate Peptide 2: 3.6 μM) and p70S6K kinase (20nM). After 15 min, the reaction was initiated by adding ATP to a final concentration of 100 μM, including approximately 0.4 μCi/μL γ-P32 ATP. Reactions were performed in a total volume of 25 μL. After 30 minutes, the reaction was terminated by spotting 17.5 μL on a circular P81 phosphocellulose paper (diameter 2.1 cm, Whatman) followed by washing four times (five minutes each wash) with 0.75% phosphoric acid and once with acetone. The dried P81 papers were transferred to a scintillation vial with 5 ml of scintillation cocktail and the counts per minute (CPM) determined with a Beckmann 6000 scintillation counter. IC50 values were defined to be the concentration of inhibitor at which the CPM was 50% of the control sample, corrected by the background.
The human melanoma cell line (451Lu) was isolated as previously described 30 and cultured in Dulbecco’s modified Eagle’s medium supplemented with 5% fetal bovine serum. Normal human primary melanocytes (FOM) and fibroblasts (FF) were isolated from the human epidermis of neonatal foreskins and cultured as described 31. Cells (5000/well) were seeded in 96-well plates and allowed to adhere overnight before treatment with 40 or the DMSO vehicle control for 72 h. Cells were then directly incubated with MTS substrate (CellTiter-96 Aqueous One Solution Cell Proliferation Assay, Promega). Absorbance was measured at 490 nm as per the supplier’s instructions and percent proliferation was normalized to the absorbance of DMSO-treated cells. For each experiment, cell line, and treatment, the absorbance values of at least 4 wells were used for data analysis and the experiment was conducted in triplicate.
Proteins were extracted and immunoblotted as previously described 32. Antibodies to phospho-MEK, total-MEK, phospho-ERK, total-ERK, and HSP90 were from Cell Signaling Technology (Beverly, MA).
We would like to thank Jessie Villanueva for useful discussions. This work was supported by NIH grant CA114046 awarded to M.H, R.M. and J.W. This project was also funded in part with Federal funds from the National Cancer Institute’s Initiative for Chemical Genetics, National Institutes of Health, under Contract No. N01-CO-12400 with some experiments performed with the assistance of the Chemical Biology Platform of the Broad Institute of Harvard and MIT. We acknowledge support of the Protein Expression and Libraries core facility at the Wistar Institute, which is supported by NIH grant CA010815.