Supplementary Information

Transcription

1 1 Supplementary Information A unique inhibitor binding site in ERK1/2 is associated with slow binding kinetics Apirat Chaikuad, Eliana Tacconi, Jutta Zimmer, Yanke Liang, Nathanael S. Gray, Madalena Tarsounas, Stefan Knapp Supplementary Table 1: Crystallographic data collection and refinement Page 2 statistics Supplementary Figure 1: Structural comparison between ERK-SCH Page 3 complex and unphosphorylated, inactive and phosphorylated, active ERK structures Supplementary Figure 2: Selectivity of SCH across the human Page 4 kinome using KINOMEScan assays. Supplementary Figure 3: Enzyme kinetic data for the interaction of Page 5 SCH with ERK1/2 and its most relevant off targets. Supplementary Figure 4: Overview of the interactions of SCH with Page 6 haspin and JNK1 Supplementary Figure 5: Kinetic binding assays of SCH and VTX- Page 7 11e on ERK1/2 Supplementary Figure 6: Binding of 5-iodotubercidin to ERK1/2 and haspin Page 8 Supplementary Figure 7: Supplementary Figure 8: Supplementary Figure 9: Supplementary Figure 10: Supplementary Figure 11: Supplementary Figure 12: Supplementary Figure 13: Supplementary Notes Structural comparison of the ERK-SCH complex with the inactive and active ERKs Structural evaluation of a possible existence of the P-loop pocket in some off-targets. Structure-based sequence alignment of ERK1/2 and SCH off-target kinases Binding kinetics of SCH to the off-target kinases Effects of SCH towards BRCA-deficient MDA-MB 231 cells Effects of ERK and PARP inhibitors on BRCA2- deficient cells Uncut gel images Synthesis and characterization of SCH derivatives Page 10 Page 11 Page 12 Page 13 Page 13 Page 15 Page 16 Page 18

2 2 Supplementary Results Supplementary Table 1: Data collection and refinement statistics ERK1-SCH ERK2-SCH haspin-sch JNK1-SCH ERK2- VTX-11e Data collection Space group P 2 1 P P P P Cell dimensions a, b, c (Å) 61.98, 94.01, , 75.54, , 78.36, , 71.55, , 96.73, α, β, γ ( ) 90.0, 91.7, , 90.0, , 90.0, , 90.0, , 90.0, Resolution (Å) ( )* ( ) ( ) ( ) ( ) R merge (0.618) (0.710) (0.693) (0.497) (0.652) I /σi 8.8 (2.2) 12.5 (2.6) 17.9 (2.4) 14.9 (3.1) 15.3 (3.0) Completeness (%) 95.8 (93.3) 99.9 (99.7) (100.0) 99.9 (99.4) 98.2 (96.9) Redundancy 4.1 (3.9) 7.9 (7.5) 6.1 (6.3) 6.4 (4.9) 7.1 (7.1) Refinement Resolution (Å) No. reflections 140,419 (19,971) 60,441 (8,701) 99,002 (14,277) 64,041 (9,131) 81,004 (11,574) R work / R free 0.147/ / / / / No. atoms Protein 5,916 2,803 2,738 2,985 2,941 SCH772984/ VTX-11e Water and solvents B factors (Å 2 ) Protein SCH772984/VTX-11e Water and solvents r.m.s. deviations Bond lengths (Å) Bond angles ( ) *Values in parentheses are for highest-resolution shell.

3 3 Supplementary Figure 1 Supplementary Figure 1: Structural comparison of inactive unphosphorylated ERK (left, pdb id: 1ERK) and active phosphorylated ERK (right, pdb id: 2ERK). Shown is a surface representation of the SCH ERK complex with the bound inhibitor in yellow stick overlaid. The insets show the essential alterations of the P-loop and αc required for the accommodation of the inhibitor within the kinase.

4 4 Supplementary Figure 2 Supplementary Figure 2: Selectivity of SCH across the human kinome using KINOMEScan assays. Competition binding using the inhibitor at 1µM concentration reveals high selectivity towards ERKs.

5 5 Supplementary Figure 3 Supplementary Figure 3: Enzyme kinetic data for the interaction of SCH with ERK1/2 and its most relevant off targets. The dose response titration curves are generated using the ATP-competitive assay platforms from life technology TM ( IC 50 values shown in the table on the right and represent means of three technical repeats.

6 6 Supplementary Figure 4 Supplementary Figure 4: Overview of the interactions of SCH with haspin and JNK1. Surface representations visualizing the binding pockets of the inhibitor within the ATP binding site of the kinase haspin (a) and JNK1 (b). The main structural elements are labelled. The below panels show a 2F O - F C omitted electron density map contoured at 1σ. Noted that additional electron density was present in the JNK1 binding pocket that was interpreted as a β,γ-imido triphosphate group and a magnesium ion, the remnant of the hydrolysis of AMP-PNP which was present in the initial co-crystallization. (c) Derivatives of SCH The inhibitor has been modified at the R1 position as indicated in the table. Data represent temperature shift values (averaged values of two replicates).

7 7 Supplementary Figure 5 Supplementary Figure 5: Kinetic (BLI) and thermodynamic (ITC) binding assay data for the binding of SCH and VTX-11e to ERKs. Isothermal titration calorimetry (ITC) for the inhibitor with ERK1 at 15 C (a) and for ERK2 at 30 C (b). Note that the non-linear least squares fit using a single binding site model does not represent well the measured data potentially due to non-equilibrium kinetic effects of inhibitor binding and this is indicated by arrows. c) Thermodynamic binding parameter determined by a non-linear least squares fit to a single site model. d) Bio-Layer Inferometry (BLI) for the binding kinetics of ERK1 with SCH e) BLI data for the interaction between ERK1 with VTX-11e reveal also a similar slow kinetic pattern. f) Fitted binding kinetic parameters.

8 8 Supplementary Figure 6a Supplemental Figure 6a: Binding of 5-iodotubercidin to ERK1/2 and haspin. Bio-Layer inferometry (BLI) (a) and isothermal titration calorimetry (ITC) (b) analyses of the inhibitor binding to the studied kinases. For BLI, the insets show steady state fitting, while for ITC, the insets show raw titration data. c) The summary of kinetic constants from the two measurements. d) Structural insights into accommodation of 5-iodotubercidin within haspin (left, pdb id: 4OUC) and ERK1 (right, pdb id: 2ZOQ) reveal the lack of the aromatic gatekeeper stacking with the inhibitor iodine in ERK1, a potential key for the slow off-rate for its type-i interaction in haspin.

9 9 Supplementary Figure 6b Supplementary Figure 6b: Inhibition of ERK signalling by FR after inhibitor washout. a) BLI data for the binding of FR to ERK2. b) Cellular activities FR towards inhibition of ERK/MAPK signalling in human MDA-MB 231 cells. Cells were treated with 500 nm for FR180204, for 4 hours and lysates were analysed using Western blotting before and after inhibitor wash out. Western blots band intensities were quantified and normalized relative to DMSO-treated cells and loading controls. Prolonged inhibition was not detected for FR for which the ERK activity recovered within half an hour, consistent with the fast off-rate of this inhibitor measured in vitro. Quantification of Western blot signals for indicated phospho-proteins relative to loading controls is shown in the extrapolated recovery curves (lower panels) and reflects the recovery rates of MAPK pathway specific phosphorylation events. Uncut images of the exposed films used to generate the figure are shown on the right. Western Blots have been repeated twice with identical results.

10 10 Supplementary Figure 7 Supplementary Figure 7: Structural comparison of the ERK-SCH complex with the inactive and active ERKs. Superimposition of ERK2-SCH with inactive and active ERKs (left, pdb-id: 1ERK and 2ERK, respectively) and ERK1-SCH with monophosphorylated ERK1 (right, pdb-id: 2ZOQ) reveals that while most parts of the kinase are highly static, slight structural alterations are evident for regions in proximity to the inhibitor binding pocket, including the P-loop site, and the activation segment αef/αf loop, MAPK-specific insertion, L16 loop and C-terminal helix. Alterations of structural elements important for ERK-MEK and ERK dimerization are likely to provide a structural rational for the lack of ERK phosphorylation and activation in the presence of SCH

11 11 Supplementary Figure 8 Supplementary Figure 8: Structural evaluation for a possible existence of the P-loop pocket in SCH off-targets. Surface representation of the ATP binding pockets of some of the off-target kinases with SCH from the ERK complexes in yellow stick overlaid. The pdb ids for each structure used are in brackets. The conformations of the residues at the tip of the P-loop of each structure are annotated, and are referred to as in and out arrangements for the side chains pointing inward or outward to the ATP binding pocket, respectively.

12 12 Supplementary Figure 9 Supplementary Figure 9: Structure-based sequence alignment of ERK1/2 and SCH off-target kinases. Sequences shown are for the regions relevant for the inhibitor binding with potential key residues for the interaction highlighted.

13 13 Supplementary Figure 10 Supplementary Figure 10a: Binding kinetics of SCH to off-target kinases. Bio- Layer inferometry (BLI) data for eight off-target kinases, all of which demonstrated fast associations and dissociations of the inhibitor unlike the slow kinetic pattern observed for ERKs. The insets show steady state fitting. The determined kinetic constants are summarized in the table.

14 14 Supplementary Figure 10b: Binding of SCH and VTX-11e to ERK2 mutants. A bar diagram showing VTX-11e dissociation half lifes is shown on the left and a table compiling measured BLI data on single and double mutants is shown on the right. Supplementary Figure 11 Supplementary Figure 11: Effects of SCH on the proliferation of BRCA2- deficient MDA-MB 231 cells. Western blots of extracts prepared from MDA-MB 231 cells demonstrating the sustained knockdown of BRCA2 by doxycycline (DOX)-induced shrna and suppression of ERK1/2 activation following SCH exposure for three days. SMC1 (Structural maintenance of chromatin) was used as a loading control. Effects on cell doubling times in the presence and absence of DOX are show in the lower panel. Data represent means of two independent experiments.

15 15 Supplementary Figure 12 Supplementary Figure 12: Effects of ERK and PARP inhibitors on BRCA2-deficient cells. Shown is a Western blot demonstrating the knock-down of BRCA2 using sirna at day 0 and day 6 and differences in cell survival rates of BRCA2 proficient and knock down cells in the presence of SCH772984, VTX-11e and olaparib. Data are representative of two independent experiments, each set up in triplicate. Error bars represent standard deviation from a single experiment.

16 16 Supplementary Figure 13 Supplementary Figure 13a: Uncut images. Shown are exposed uncut films used to generate figure 4

17 17 Supplementary Figure 13b: Uncut images. Shown are exposed uncut films used to generate figure 6

18 18 Supplementary Notes Commercial inhibitors: SCH was purchased from Selleckchem (Cat. No. S7101). VX11e was purchased from ChemieTek (Cat. No. CT-VX11e). FR was purchased from Tocris (Cat. No 3706). 4-(4-Pyrimidin-2-yl-phenyl)-piperazine-1-carboxylic acid tert-butyl ester. A solution of 4- (4-Bromophenyl)-piperazine-1-carboxylic acid tert-butyl ester (2.5 g, 7.33 mmol) in THF (20 ml) was cooled to -78 C under a dry Ar atmosphere. A 1.6 M solution of n-buli in hexanes (10 ml, 16.1 mmol) was added dropwise and stirred for 30 min. Trimethyl borate (1.63 ml, 14.7 mmol) was added dropwise. The reaction mixture was allowed to warm to room temperature gradually and stirred overnight. A saturated aqueous NH 4 Cl solution (20 ml) was added and the reaction mixture was stirred for 5 min. 85% o-phosphoric acid (2 g) was added and the reaction mixture was stirred for 1 hr. The reaction mixture was extracted with EtOAc (3 x), washed with brine, dried (Na 2 SO 4 ), filtered and concentrated to give crude boronic acid, which was used directly without purification. The resulting boronic acid was dissolved in a 1:1 (v/v) mixture of DMF/water (50 ml). 2-Chloropyrimidine (1.05 g, 9.16 mmol) was added, followed by K 2 CO 3 (6.07 g, 44.0 mmol). The mixture was degased and purged with Ar. Pd(dppf) 2 CI 2 (600 mg, mmol) was added and the reaction mixture was stirred at 80 C for 8 hr. The mixture was cooled and diluted with water and extracted with EtOAc (2 x), dried (Na 2 SO 4 ), filtered and concentrated. The crude product was purified by flash chromatography on silica gel (0-60% EtOAc in hexane) to obtain the product as a white solid (1.95 g, 78% over two steps). 1 H NMR (400 MHz, CDCl 3 ): δ 8.65 (d, J = 5.2 Hz, 2H), 8.27 (d, J = 9.2 Hz, 2H), 7.00 (t, J = 5.2 Hz, 1H), 6.90 (d, J = 9.2 Hz, 2H), 3.52 (t, J = 5.2 Hz, 4H), 3.21 (t, J = 5.2 Hz, 4H), 1.42 (s, 9H); MS m/z: [M+1]. Methyl (R)-1-(2-(tert-butoxy)-2-oxoethyl)pyrrolidine-3-carboxylate. R-Pyrrolidine-3- carboxylic acid methyl ester (1.0 g, 6.0 mmol) was dissolved in DMF (30 ml). DIEA (3.2 ml, 18.0 mmol) was added, followed by Cs 2 CO 3 (2.93 g, 9.0 mmol). tert-butyl bromoacetate (0.98 ml, 6.6 mmol) was added dropwise and the reaction mixture was stirred for 1 hr. The

19 19 reaction mixture was diluted with water and extracted with EtOAc (3 x), dried (Na 2 SO 4 ), filtered and concentrated. The crude product was purified by flash chromatography on silica gel (0-70% EtOAc in hexane) to obtain the product as a white solid (1.37 g, 94%). 1 H NMR (400 MHz, CDCl 3 ): δ 3.65 (s, 3H), 3.24 (d, J = 16.8 Hz, 1H), 3.17 (d, J = 16.8 Hz, 1H), (m, 2H), (m, 1H), (m, 1H), 2.55 (dd, J = 8.4, 8.4 Hz, 1H), (m, 2H), 1.43 (s, 9H); MS m/z: [M+1]. Methyl (R)-1-(2-oxo-2-(4-(4-(pyrimidin-2-yl)phenyl)piperazin-1-yl)ethyl)pyrrolidine-3- carboxylate. Methyl (R)-1-(2-(tert-butoxy)-2-oxoethyl)pyrrolidine-3-carboxylate (1.13 g, 4.66 mmol) was dissolved in a 1:1 (v/v) mixture of TFA and DCM (10 ml) and stirred at rt for 2 hr. The solution was concentrated to give crude (R)-2-(3-(methoxycarbonyl)pyrrolidin-1-yl)acetic acid, which was used without purification. 4-(4-Pyrimidin-2-yl-phenyl)-piperazine-1-carboxylic acid tert-butyl ester (1.32 g, 3.88 mmol) was dissolved in DCM (7 ml) and a 4 M solution of HCl in dioxane (7 ml) was added and stirred at rt for 2 hr. The reaction mixture was concentrated to give crude 2-(4-(piperazin-1- yl)phenyl)pyrimidine, which was used without purification. The resulting (R)-2-(3-(methoxycarbonyl)pyrrolidin-1-yl)acetic acid and 2-(4-(piperazin-1- yl)phenyl)pyrimidine were dissolved in DMF (20 ml). 1-Hydroxybenztriazole (HOBT) (630 mg, 4.66 mmol), 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDCI) (970 mg, 5.00 mmol) and TEA (2.8 ml, 20 mmol) were added and stirred at rt overnight. The reaction mixture was diluted with EtOAc and washed with sat. NaHCO 3, water, and brine. The organic phase was dried (Na 2 SO 4 ), filtered and concentrated. The crude was purified by flash chromatography on silica gel (0-15% MeOH in DCM) to obtain the product as a yellow solid (1.21 g, 76% over two steps). 1 H NMR (400 MHz, CD 3 OD): δ 8.74 (d, J = 4.8 Hz, 2H), 8.27 (d, J = 8.8 Hz, 2H), 7.24 (t, J = 4.8 Hz, 1H), 7.06 (d, J = 8.8 Hz, 2H), 4.38 (s, 2H), (m, 2H), 3.75 (s, 3H), (m, 2H), (m, 2H), (m, 2H), (m, 2H), 3.36 (t, J = 5.2 Hz, 2H), 3.30 (m, 1H), (m, 1H), (m, 1H); MS m/z: [M+1].

20 20 (R)-N-(3-Bromo-1H-indazol-5-yl)-1-(2-oxo-2-(4-(4-(pyrimidin-2-yl)phenyl)piperazin-1- yl)ethyl)pyrrolidine-3-carboxamide. Methyl (R)-1-(2-oxo-2-(4-(4-(pyrimidin-2- yl)phenyl)piperazin-1-yl)ethyl)pyrrolidine-3-carboxylate (410 mg, 1.0 mmol) was dissolved in MeOH (5 ml). A 1 M LiOH solution (1.2 ml) was added and stirred at rt overnight. The reaction was neutralized with 1 M HCl solution (1.2 ml) and concentrated to give crude acid as a yellow solid, which was used directly without further purification. The resulting acid and 3-bromo-1H-indazol-5-amine (424 mg, 2.0 mmol) were dissolved in DMF (5 ml). DIEA (1.74 ml, 10 mmol) was added, followed by HATU (760 mg, 2.0 mmol), and stirred at rt overnight. The reaction mixture was diluted with EtOAc and washed with water and brine, dried (Na 2 SO 4 ), and concentrated. The crude was purified by flash chromatography on silica gel (0-15% MeOH in DCM) to obtain the product as a yellow solid (301 mg, 43% over two steps). 1 H NMR (400 MHz, DMSO-d 6 ): δ (s, 1H), (s, 1H), 8.78 (d, J = 4.4 Hz, 2H), 8.24 (d, J = 9.2 Hz, 2H), 8.04 (s, 1H), 7.47 (m, 2H), 7.27 (t, J = 4.8 Hz, 1H), 7.03 (d, J = 9.2 Hz, 2H), 3.67 (m, 2H), 3.61 (m, 2H), 3.43 (m, 2H), (m, 4H), (m, 1H), (m, 1H), (m, 2H), (m, 1H), 2.03 (q, J = 7.2 Hz, 2H); MS m/z: [M+1]. (R)-1-(2-oxo-2-(4-(4-(pyrimidin-2-yl)phenyl)piperazin-1-yl)ethyl)-N-(3-phenyl-1Hindazol-5-yl)pyrrolidine-3-carboxamide. (R)-N-(3-Bromo-1H-indazol-5-yl)-1-(2-oxo-2-(4-(4- (pyrimidin-2-yl)phenyl)piperazin-1-yl)ethyl)pyrrolidine-3-carboxamide (10 mg, mmol), phenylboronic acid (3.5 mg, mmol), XPhos (1.4 mg, mmol) were mixed in dioxane (0.8 ml). Saturated Na 2 CO 3 (0.2 ml) was added and the mixture was degassed and purged with Ar. Pd(dppf) 2 Cl 2 (1.2 mg, mmol) was added and stirred at 80 C for 8 hr. The reaction mixture was cooled and diluted with a 4:1 solution of CHCl 3 /i-proh, and washed with water and brine. The organic layer was dried (Na 2 SO 4 ), filtered and concentrated. The crude was purified by HPLC (85%-5% water in MeOH with 0.5% TFA) to give the product as a TFA salt (6.3 mg, 63%).

21 21 1 H NMR (400 MHz, DMSO-d 6 ): δ (s, 1H), (d, J = 14.4 Hz, 1H), (m, 1H), 8.78 (d, J = 4.8 Hz, 2H), 8.48 (s, 1H), 8.26 (d, J = 8.8 Hz, 2H), 7.89 (d, J = 7.6 Hz, 1H), (m, 4H), 7.39 (t, J = 7.2 Hz, 1H), 7.29 (t, J = 4.8 Hz, 1H), 7.07 (d, J = 9.2 Hz, 2H), 4.54 (m, 1H), (m, 1H), (m, 4H), 3.52 (m, 2H), 3.41 (m, 4H), 3.34 (m, 2H), (m, 1H), (m, 1H), (m, 1H); 13 C NMR (100 MHz, DMSO-d 6 ): δ 170.5, 170.3, 164.1, 163.8, 157.9, 152.5, 139.2, 134.2, 133.2, 129.3, 128.1, 127.9, 127.0, 120.7, 120.3, 119.1, 115.0, 111.4, 110.3, 56.8, 55.7, 47.4, 47.3, 44.3, 42.9, 28.3; MS m/z: [M+1]. (R)-N-(3-(4-cyanophenyl)-1H-indazol-5-yl)-1-(2-oxo-2-(4-(4-(pyrimidin-2- yl)phenyl)piperazin-1-yl)ethyl)pyrrolidine-3-carboxamide. (R)-N-(3-Bromo-1H-indazol-5- yl)-1-(2-oxo-2-(4-(4-(pyrimidin-2-yl)phenyl)piperazin-1-yl)ethyl)pyrrolidine-3-carboxamide (10 mg, mmol), 4-cyanophenylboronic acid (4.2 mg, mmol), XPhos (1.4 mg, mmol) were mixed in dioxane (0.8 ml). Saturated Na 2 CO 3 (0.2 ml) was added and the mixture was degassed and purged with Ar. Pd(dppf) 2 Cl 2 (1.2 mg, mmol) was added and stirred at 80 C for 8 hr. The reaction mixture was cooled and diluted with a 4:1 solution of CHCl 3 /i-proh, and washed with water and brine. The organic layer was dried (Na 2 SO 4 ), filtered and concentrated. The crude was purified by HPLC (85%-5% water in MeOH with 0.5% TFA) to give the product as a TFA salt (4.6 mg, 45%). 1 H NMR (400 MHz, DMSO-d 6 ): δ (s, 1H), (d, J = 12.4 Hz, 1H), (m, 1H), 8.74 (d, J = 4.8 Hz, 2H), 8.47 (s, 1H), 8.21 (d, J = 9.2 Hz, 2H), 8.05 (d, J = 8.4 Hz, 2H), 7.94 (d, J = 8.4 Hz, 2H), 7.56 (d, J = 9.2 Hz, 1H), 7.50 (dd, J = 8.4, 2.0 Hz, 1H), 7.24 (t, J = 4.8 Hz, 1H), 7.02 (d, J = 8.8 Hz, 2H), 4.50 (m, 1H), (m, 1H), (m, 1H), (m, 1H), 3.62 (m, 4H), 3.36 (m, 4H), 3.29 (m, 2H), (m, 1H), (m, 1H), (m, 1H); 13 C NMR (100 MHz, DMSO-d 6 ): δ 170.7, 170.4, 164.1, 163.8, 157.9, 152.5, 139.2, 138.8, 133.9, 133.4, 129.4, 127.9, 127.2, 121.1, 120.4, 119.4, 119.1, 115.0, 111.8, 110.3, 109.9, 56.8, 55.7, 47.4, 47.3, 44.3, 42.9, 28.2; MS m/z: [M+1].