Differential Kinobeads Profiling for Target ... - ACS Publications

Subscriber access provided by University of Colorado Boulder

Letter

Differential kinobeads profiling for target identification of irreversible kinase inhibitors Lars Dittus, Thilo Werner, Marcel Muelbaier, and Marcus Bantscheff ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.7b00617 • Publication Date (Web): 06 Sep 2017 Downloaded from http://pubs.acs.org on September 7, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Letter to ACS Chemical Biology

Differential kinobeads profiling for target identification of irreversible kinase inhibitors Lars Dittus, Thilo Werner, Marcel Muelbaier* and Marcus Bantscheff* Cellzome GmbH, A GlaxoSmithKline company, Meyerhofstrasse 1, D-69117 Heidelberg, Germany

*) For correspondence: [email protected] and [email protected]

Keywords: targeted covalent kinase inhibitor, intracellular target engagement, differential kinobeads profiling, covalent vs non-covalent off-target binding Declaration: The authors are employees of Cellzome GmbH and GlaxoSmithKline. The company funded the work.

Abstract Chemoproteomics profiling of kinase inhibitors with kinobeads enables the assessment of inhibitor potency and selectivity for endogenously expressed protein kinases in cell lines and tissues. Using a small panel of targeted covalent inhibitors, we demonstrate the importance of measuring covalent target binding in live cells. We present a differential kinobeads profiling strategy for covalent kinase inhibitors where compound is added either to live cells or to cell extract that enables the comprehensive assessment of inhibitor selectivity for covalent and non-covalent targets. We found that Acalabrutinib, CC-292 and Ibrutinib potently and covalently bind TEC family kinases but only Ibrutinib also potently binds to BLK. ZAK was identified as a sub micromolar affinity Ibrutinib offtarget due to covalent modification of Cys22. In contrast to Ibrutinib, 5Z-7-Oxozeaenol reacted with Cys150 next to the DFG loop, demonstrating an alternative route to covalent inactivation of this kinase e.g. to inhibit canonical TGF-β dependent processes.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 15

The clinical success of Ibrutinib (Imbruvica), a covalent inhibitor of Bruton’s Tyrosine Kinase (BTK) for the treatment of B-cell malignancies1 triggered renewed interest in irreversible kinase inhibitors throughout academic and industrial research2. Targeted covalent kinase inhibitors (TCIs) constitute of a rationally designed small molecule that was optimized for potent and selective binding to the target protein and a low-reactive electrophile that is designed to react with a cysteine residue within or close to the binding site3. Thus, long lasting and efficient inhibition of target kinases is achieved by non-equilibrium binding and because the effect duration predominantly depends on the turnover of the target protein rather than the pharmacokinetic properties of the drug. However, covalent drugs also raise concerns about polypharmacology, non-specific, irreversible protein modification and idiosyncratic toxicity2, 4-6. Chemoproteomics approaches probe the selectivity of small molecule inhibitors against the proteome complementing conventional profiling approaches that are limited by the typically few hundred assays available in the respective assay panels7-10. Recently, target engagement of covalent kinase inhibitors was detected in living cells by combining cell-permeable activity-based probes and bioorthogonal click chemistry. Proteome-wide analyses identified covalently modified off-targets including non-kinases and uncharacterized proteins11-13. Such chemoproteomics approaches require the design and synthesis of suitable affinity-based probes for the small molecule under investigation and the position of the enrichment handle of these probes introduces a bias towards certain binding modes11. Target class-specific probes such as e.g. kinobeads7 (Fig. S1), allow assessing the selectivity of compounds only within the captured sub-proteome (e.g. kinases) but do not require such bespoke probes. It is less clear, however, how well lysate-based assays reflect intracellular kinase selectivity for targeted covalent inhibitors since cysteine reactivity depends on the native redox environment14 and local pH15,

16

and lysis can cause chemical lability of cysteine PTMs (nitrosation, palmitoylation,

prenylation, and Michael addition to oxidized lipids)14 and promotes rapid oxidation17 of sulfhydryl groups. To address this question, we compared target engagement of three targeted covalent inhibitors of BTK in a cell-based and a cell extract-based competition binding assay with kinobeads. Primary B-cells and cell extracts derived from primary B-cells were incubated with Ibrutinib (compound 1), CC-29218 (compound 2), and QL4719 (compound 3) over a range of concentrations. Proteins not bound by the covalent inhibitors were affinity enriched with kinobeads, eluted, and BTK concentrations were quantified by immunoblot analysis. This enabled the determination of binding curves by fourparameter non-linear regression curve fitting based on the enrichment of non-competed BTK as described previously20. From such binding curves, we derived inhibitor concentrations for which binding to the kinobeads was reduced by 50 % (IC50). For convenience, these concentrations are 2 ACS Paragon Plus Environment

Page 3 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


referred to as pIC50 defined as –log10 [IC50]. In the lysate-based assay similar IC50s were determined for all three tested compounds (pIC50,Ibrutinib = 6.7±0.1, pIC50,CC-292 = 6.9±0.1, and pIC50,QL47 = 6.5±0.1, Fig. 1A, Fig. S2, Tab. S1). When compounds were applied on live cells, however, significant differences in BTK engagement for the inhibitors were determined with IC50s spanning nearly two orders of magnitude between binding by Ibrutinib (pIC50 = 8.8±0.0) and binding by QL47 (pIC50 = 7.0±0.2). The observed rank order agreed well with results of a BTK-dependent calcium mobilization assay suggesting that the cell-based kinobeads assay was more predictive. In a next step, we sought to exploit differences in target binding of covalent inhibitors in cell-based and lysate-based experiments to distinguish covalent from non-covalent kinase targets. Using Ibrutinib and a saturated analogue containing a non-reactive propionamide instead of an acrylamide moiety but otherwise identical chemical structure (compound 4) as model systems experimental conditions for the cell-based kinobeads assay were optimized to achieve strong covalent target binding of Ibrutinib down to low nanomolar concentrations (Fig. 1B, Fig. S3) and efficient wash-out of non-covalently bound kinase inhibitor (Fig. 1C, Fig. S4). We hypothesized that comparison of dosedependent competition binding experiments where live cells were incubated with compound prior to lysis and incubation with kinobeads with lysate-based experiments would identify non-covalent targets of kinase inhibitors by an IC50 shift in cell based experiments to lower affinities due to reequilibration of drug target interactions after cell lysis. In contrast, covalent drug-target interactions would not be affected by cell lysis and dilution of the protein extracts and would hence appear equipotent or somewhat more potent in cell-based experiments due to more efficient target engagement in cells (Fig. 1D). We then generated cell-based and lysate-based kinobeads profiles for Ibrutinib in the B lymphocyte cell line Ramos. After lysis of the compound treated cells, kinobeads-enriched kinases were analyzed by quantitative mass spectrometry (Tab. S2A). Compound concentration dependent displacement from the kinobeads and binding affinities were determined for a total of 1900 proteins including 209 protein kinases (Tab. S3). When preparing cell extracts for kinobeads pulldowns compound and protein concentrations were diluted 7.6-fold (log10 = 0.88). This dilution factor (Tab. S4) was ignored for IC50 calculation in cell-based profiling samples. Hence, for non-covalent targets of the tested inhibitors, re-equilibration is expected to occur resulting in an approximately 8-fold potency shift whilst no such shift is expected for targets that are irreversibly bound to the tested inhibitors. The TEC family kinases BTK and TEC as well as BLK were identified as the most potently bound Ibrutinib target kinases, additional sub micromolar kinase targets include RIPK2 and 3, FGR and SRC family kinases (Fig. 1E). When comparing cell-based and lysate-based profiling experiments, many kinases displayed the expected 8-fold IC50 drop comparing cell-based and lysate-based potency, suggesting that these kinases were non-covalent targets of the drug. In contrast, for BTK, TEC, BLK and ZAK 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 15

similar or higher pIC50 values were obtained in cell-based experiments suggesting covalent binding to Ibrutinib. TEC family kinases as well as the SRC kinase BLK contain a cysteine residue at the hinge 6 position which is targeted by Ibrutinib3. ZAK has been suggested as an additional covalent target in a recent report11. Ibrutinib profiling in epithelial A549 cells suggested EGF-receptor as an additional covalent target albeit with lower potency. Covalent binding can be rationalized by reaction to Cys847 that corresponds to Cys481 of BTK. In kinobeads profiling experiments with Ibrutinib, we robustly identified 247 kinases in total of which 5 were engaged irreversibly, and 17 were indicated as noncovalent binders with sub micromolar IC50 (Tab. S5). In order to validate our approach additional kinobeads profiling experiments were performed with a saturated analogue of Ibrutinib 4. As expected all kinase targets displayed the apparent affinity drop corresponding to the dilution factor (Tab. S4) when preparing lysates from live cells (Fig. 1F).

Figure 1: Setting up differential kinobeads selectivity profiling to distinguish covalent and reversible targets of covalent kinase inhibitors. (A) Comparison of targeted covalent BTK inhibitors Ibrutinib, CC-292, and QL47 in different assays reveals differences in target engagement measured in cells and in lysate using a kinobeads competition binding assay. The comparison to a functional assay reporting inhibition of BCR signaling by modulation of calcium mobilization indicates cellular kinobeads assays being more predictive for the functional effect of covalent kinase inhibitors than lysate-based assays. The chart represents aggregated pIC50 values afforded from ≥ 2 replicates in the respective assay. Error bars indicate standard errors of the mean (SEM). (B) Determination of target engagement in the dynamic range of reaction kinetics. Increasing occupancy of BTK was observed for the reactive compound. At low nanomolar concentrations of Ibrutinib 50 % occupancy was achieved after 1 h of cell treatment and no time-dependent increase in BTK occupancy was observed for the non-reactive analogue. The average of the BTK signals and SEM was calculated from two replicates. (C) The dissociation of compound reversibly bound to proteins was supported by intensive cell washing (2x incubation in fresh medium for 30 min). The immunoblot shows the BTK signals after enrichment of proteins using kinobeads and elution of the captured proteins. (D) Concept of differential kinobeads profiling. Covalent and reversible targets of reactive kinase inhibitors can be distinguished by different target engagement measured in living cells and cell extracts. The afforded curves represent kinase binding to kinobeads competed by increasing concentrations of a kinase inhibitor. (E) Kinases robustly identified to be inhibited with pIC50 ≥ 6 by Ibrutinib and (F) saturated Ibrutinib as determined by differential kinobeads profiling. Scattered lines represent the calculated dilution-factors due to cell lysis. All values are average values from 2 replicates.


Page 5 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


We then applied our differential kinobeads profiling strategy to characterize target kinases for a small panel of covalent kinase inhibitors in Ramos and A549 cells: the covalent BTK inhibitors CC29218 and Acalabrutinib21 (compound 5), as well as 5Z-7-Oxozeaenol (compound 6), a natural product of fungal origin22,

23

. Comparison of lysate-based and cell-based kinobeads profiles for CC-292

suggested 3 of the 231 quantified kinases as covalent binders. These include the previously described covalent targets TEC and BTK18 (Fig. 2A) as well as BLK. In contrast to Ibrutinib, CC-292 was substantially less potently binding to BLK demonstrating superior selectivity for covalent attachment to the hinge 6 position of TEC family kinases. For JAK3 and LATS1 we determined low affinity binding in lysate but not in cells suggesting both as reversible targets of CC-292. Similar to CC-292, also Acalabrutinib displayed superior kinase selectivity as compared to Ibrutinib in concordance with a recent study.21 Amongst the 236 identified kinases the only cellular targets that were determined to be covalently bound were BTK and TEC with pIC50 values of 8.1 and 7.8, respectively. However, in contrast to recently reported data21,

24

, these analyses revealed cross-

reactivity of the compound with endogenous TEC. Further, LIMK1, RIPK2 and RIPK3 showed valid target engagement in lysate but not in the cell-based kinobeads assay suggesting those kinases as reversible binders of Acalabrutinib (Fig. 2B). For 5Z-7-Oxozeaenol 18 kinases were identified as sub micromolar targets of which 8 kinases were covalently bound by the compound (Fig. 2C). Covalent kinase targets include the 3 previously identified covalent 5Z-7-Oxozeaenol targets BMP2K, GAK, MAP3K722, 23 as well as the novel covalent targets MKNK1, ZAK, CDKL5, and MAP2K2. All of these kinases contain a cysteine directly preceding the DFG motif (DFG-1 position) that was previously identified as covalent binding site of 5Z-7Oxozeaenol22, 23 underscoring the plausibility of our data. In addition, 14 kinases were bound in a predominantly non-covalent manner. These included a number of kinases containing a conserved cysteine residue at the DFG-1 position which apparently displayed little or no reactivity towards the compound.

Figure 2: Kinase targets of the covalent kinase inhibitors CC-292 (A), Acalabrutinib (B), and 5Z-7-Oxozeaenol (C) as identified by differential kinobeads selectivity profiling. Out of the approximately 200 protein kinases robustly identified in cell-based (red) and cell extracts-based (blue) experiments only those kinases are shown for which a pIC50 value > 6 was determined. Differential kinobeads profiling identified TEC and BTK as potent covalent targets of CC-292 and Acalabrutinib and BLK as sub micromolar covalent target of CC-292. JAK3 and LATS1 as well as LIMK1, RIPK2 and RIPK3 were identified as reversible binders of CC-292 and Acalabrutinib, respectively. 5Z-7-Oxozeaenol showed a very different selectivity profile for both, covalent and reversible targets. All pIC50 values are average values from 2 replicates and error bars represent the standard error of the mean.

To validate ZAK as covalent target of the structurally very different kinase inhibitors Ibrutinib and 5Z7-Oxozeaenol and to further investigate if both inhibitors modified the same cysteine residues we 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 15

incubated recombinantly expressed ZAK with excess amounts of both inhibitors followed by enzymatic digestion and mass spectrometry-based identification of binding sites. Cysteine residues of three peptides were identified showing mass differences corresponding to 5Z-7Oxozeaenol modification (Fig. 3A; Tab. S6). We identified a covalent modification of cysteine Cys150 which is part of the CDFG sequence (DFG-1 position), and thus, is located inside the ATP-binding site of ZAK. Reactivity of 5Z-7-Oxozeanol with this cysteine residue is in agreement with a previous report demonstrating covalent modification of CDFG sequence motifs in other kinases.25 Competition effects observed in the differential kinobeads profiling can be assigned to Cys150 binding. In addition, modification of the cysteine residues Cys231 & Cys358 was observed upon incubation with a 30-fold molar excess of 5Z-7-Oxozeaenol for 24 h. In contrast to 5Z-7-Oxozeaenol, Ibrutinib has not been shown to modify CDFG cysteines but reacts with the hinge 6 cysteine, Cys481 in BTK. ZAK does not contain a cysteine homologous to Cys481 in BTK but instead contains a serine residue (Ser89) at the corresponding position. To provide further evidence of covalent binding, two approaches were used. In a first approach, we synthesized a tagged analogue of Ibrutinib (compound 7) that was immobilized on beads and used in a competition binding experiment in Ramos cell extract. After SDS-elution of bead-bound proteins, covalently captured proteins were trypsin digested on the beads and identified by mass spectrometry (Tab. S7). In this approach, ZAK was only identified after trypsinization of the beads suggesting it was covalently bound to the beads. Besides ZAK, only the known covalent targets BTK, TEC, and BLK were significantly abundant in the on-bead digestion fraction (Fig. 3B). To identify the covalent binding site of Ibrutinib to ZAK, recombinant ZAK was incubated with Ibrutinib as described above for 5Z-7Oxozeaenol. We identified a unique peptide with a mass shift corresponding to the addition of Ibrutinib (Fig. 3C; Tab. S6) this peptide contains the cysteine residue Cys22 which was suggested as a potential covalent binding site of Ibrutinib11. Crystallography data25 revealed that Cys22 (glycine loop 1 position) is located close to the reactive moiety of Ibrutinib in similar distance as Ser89 (Fig. 3D). Since no further peptide was modified by Ibrutinib, we conclude that covalent binding of ZAK requires Ibrutinib directing its acryl amide towards Cys22 after binding to the ATP-pocket.

Figure 3: Validation of ZAK as covalent target of Ibrutinib and 5Z-7-Oxozeaenol. (A) Tandem mass spectra of peptides identified to include a covalent modification of cysteine by 5Z-7-Oxozeaenol. Mass differences between two peaks corresponding to covalently modified cysteine are indicated by arrows. (B) Confirmation of ZAK as covalent target of Ibrutinib by an orthogonal approach utilizing an Ibrutinib analogue (compound 7) immobilized on a solid bead matrix. The heatmap shows MS1 signal abundances (TOP3 method) of specifically captured proteins (i.e. competed by more than 40 % by 10 µM Ibrutinib; Fig. S5) upon SDS elution or bead trypsinization. (C) Tandem mass spectrum of a tryptic peptide containing Ibrutinib-modified Cys22. (D) Crystal structure of ZAK (green, PDB ID: 5HES25) overlaid by a crystal structure of BTK in complex with an Ibrutinib analogue (brown, PDB ID: 3GEN26). Cysteines located in the ATP binding site of ZAK and Ser89 are highlighted. The arrows indicate amino acids covalently reacted with Ibrutinib or 5Z-7-Oxozeaenol identified by mass spectrometry. The close spatial proximity of Cys22 to the hinge 6 serine enables covalent reaction of Ibrutinib with ZAK.


Page 7 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In summary, the differential kinobeads profiling strategy presented in this report confirmed known covalent targets of the tested inhibitors and identified the binding modalities for previously described off-targets as well as novel kinase off-targets. When comparing the selectivity profiles of the different covalent kinase inhibitors, our data shows only a limited number of common covalent (off-)targets indicated by equal or even slightly stronger target engagement in cells than in lysate. A noticeable difference of the tested covalent BTK inhibitors was observed for the covalent engagement of BLK. Ibrutinib and CC-292 bound to BLK with cellular pIC50 of 9.3±0.1 and 6.3±0.0, respectively, while Acalabrutinib did not engage cellular BLK at all. A recent report demonstrated the benefit of co-inhibition of BTK and BLK by Ibrutinib in the treatment of pre-BCR+ B-cell acute lymphoblastic leukemia27. It is yet to be seen if the superior selectivity of second generation covalent BTK inhibitors such as Acalabrutinib leads to better therapeutic index in the treatment of B-cell lymphoma indications.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 15

Methods See supplementary information for additional methods. Kinobeads pulldown. For a pulldown using kinobeads, the bead matrix was washed twice with 50fold excess in volume of drug pulldown (DP) buffer (50 mM Tris/HCl pH 7.4, 5 % (v/v) glycerol, 1.5 mM MgCl2, 150 mM NaCl, 1 mM Na3VO4, 25 mM NaF) and equilibrated once with 50-fold excess in volume of DP/0.4 % (v/v) NP-40. The settled beads were resuspended in DP/0.4 % (v/v) NP-40 to get a 5 % (v/v) slurry. Beads were added to the wells of 96-well or 384-well filter plates in a 1:7 ratio protein [mg] : beads [µl]. After addition of the lysate to the dried beads, incubation was done at 4 °C for 1 h overhead rotating and 900 rpm horizontal shaking for 96-well plates and 384-well plates, respectively. After protein capturing, beads were washed in 300 bead volumes DP/0.4 % (v/v) NP-40 and 150 bead volumes DP/0.2 % (v/v) NP-40. Subsequently, residual washing buffer was removed by centrifugation (314 g, 2 min, RT) and bound proteins were denatured for 30 min, 50 °C using 2X LDS sample buffer (NuPAGE #NP0007) / 50 mM DTT. Elution was done by centrifugation (314 g, 4 min, RT). LDS eluates were stored at -20 °C until further processing by Western blot or mass spectrometry analysis.

Cell-based kinobeads assay. Ramos cells or isolated human B-cells adjusted to a density of 5.6*1606 cells per ml in RPMI-1640 medium (Gibco #21875-034) supplemented with 0.1 % (v/v) FCS (Gibco #10270) were treated with 10 – 1.7 – 0.3 – 0.05 – 0.008 – 0.0013 – 0.0002 µM compound or DMSO (final DMSO concentration: 0.5 % (v/v)) for 1 h, 37 °C, 5 % (v/v) CO2, 100 rpm shaking. Per data point, 1.4*108 cells were collected and centrifuged for 4 min, 491 g, RT. The cells were washed twice by resuspending in 25 ml fresh medium and incubation for 30 min, 37 °C, 5 % (v/v) CO2, 100 rpm shaking and subsequently twice by resuspending in 10 ml ice cold PBS (Gibco #14190-094). After every washing step, cells were centrifuged for 4 min, 491 g, at RT or 4 °C prior to medium incubation or PBS resuspension, respectively. For the treatment of adherent A549 cells, 5*106 cells were seeded in 25 ml DMEM medium (Gibco #41965-039) supplemented with 10 % (v/v) FCS per 15 cm cell culture dish. After 72 h ( approximately 90 % confluency) the medium was removed and replaced by 15 ml DMEM medium/0.1 % (v/v) FCS containing 10 – 1.7 – 0.3 – 0.05 – 0.008 – 0.0013 – 0.0002 µM compound or DMSO (final DMSO concentration: 0.5 % (v/v)). Cells were incubated for 1 h, 37 °C, 10 % CO2. After incubation, the compound containing medium was removed and the cells were incubated 1x 30 min in 15 ml medium and 1x 30 min in 15 ml pre-warmed PBS. For harvesting, 3 ml versene solution was added per plate and the plates were incubated for 10 min, 37 °C, 5 % CO2, 100 rpm shaking. The cells were transferred to 15 ml tubes using 12 ml ice cold PBS. The cell suspension was centrifuged for 3 min, 491 g, 4 °C and the supernatant was removed. The cells were 8 ACS Paragon Plus Environment

Page 9 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


resuspended once more in 10 ml PBS and centrifuged for 4 min, 491 g, 4 °C. After the last washing step, suspension and adherent cells were treated equally by removing the supernatant completely and freezing the cells in liquid nitrogen to be stored at –80 °C until cell lysis. A kinobeads pulldown was performed for each individual lysate as described above.

Lysate-based kinobeads assay. Per data point 1 ml lysate (protein concentration adjusted to 5 mg ml1

at 0.4 % (v/v) NP-40) was incubated with 50 – 8.3 – 1.4 – 0.23 – 0.04 – 0.0064 – 0.0011 µM

compound or DMSO (final DMSO concentration in lysate: 0.5 % (v/v)). After incubation for 1 h at 4 °C, overhead rotating, a kinobeads pulldown was performed as described above.

Acknowledgments We would like to thank M. Klös-Hudak, K. Kammerer, and T. Rudi for assistance with sample preparation for mass spectrometry and operating LC-MS/MS instruments; J. Stuhlfauth, N. GarciaAltrieth and K. Bess for providing cell lysates; D. Thomson and A. Wagner for consultation about chemistry, R. Gujjula for synthesis of compounds and Gerard Drewes and Bernhard Kuster for their valuable feed-back and support.

Author Contributions LD designed and performed experiments. TW assisted in mass spectrometry analyses. MM and MB conceptualized the study. LD, MM and MB wrote the manuscript.

Supporting Information Available: Detailed Materials and Method section; supplementary figures, supplementary datasets. This material is available free of charge via the Internet http://pubs.acs.org.

All authors are employees and/or shareholders of Cellzome GmbH and GlaxoSmithKline. The companies funded the work.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 15

References 1. Garber, K. (2014) Kinase inhibitors overachieve in CLL, Nat Rev Drug Discov 13, 162-164. 2. Singh, J., Petter, R. C., Baillie, T. A., and Whitty, A. (2011) The resurgence of covalent drugs, Nat Rev Drug Discov 10, 307-317. 3. Honigberg, L. A., Smith, A. M., Sirisawad, M., Verner, E., Loury, D., Chang, B., Li, S., Pan, Z., Thamm, D. H., Miller, R. A., and Buggy, J. J. The Bruton tyrosine kinase inhibitor PCI-32765 blocks Bcell activation and is efficacious in models of autoimmune disease and B-cell malignancy, Proc Natl Acad Sci U S A 107, 13075-13080. 4. Garuti, L., Roberti, M., and Bottegoni, G. Irreversible protein kinase inhibitors, Curr Med Chem 18, 2981-2994. 5. Johnson, D. S., Weerapana, E., and Cravatt, B. F. (2010) Strategies for discovering and derisking covalent, irreversible enzyme inhibitors, Future Med Chem 2, 949-964. 6. Mah, R., Thomas, J. R., and Shafer, C. M. (2014) Drug discovery considerations in the development of covalent inhibitors, Bioorg Med Chem Lett 24, 33-39. 7. Bantscheff, M., Eberhard, D., Abraham, Y., Bastuck, S., Boesche, M., Hobson, S., Mathieson, T., Perrin, J., Raida, M., Rau, C., Reader, V., Sweetman, G., Bauer, A., Bouwmeester, T., Hopf, C., Kruse, U., Neubauer, G., Ramsden, N., Rick, J., Kuster, B., and Drewes, G. (2007) Quantitative chemical proteomics reveals mechanisms of action of clinical ABL kinase inhibitors, Nat Biotechnol 25, 1035-1044. 8. Duncan, J. S., Gyenis, L., Lenehan, J., Bretner, M., Graves, L. M., Haystead, T. A., and Litchfield, D. W. (2008) An unbiased evaluation of CK2 inhibitors by chemoproteomics: characterization of inhibitor effects on CK2 and identification of novel inhibitor targets, Mol Cell Proteomics 7, 1077-1088. 9. Kruse, U., Pallasch, C. P., Bantscheff, M., Eberhard, D., Frenzel, L., Ghidelli, S., Maier, S. K., Werner, T., Wendtner, C. M., and Drewes, G. (2011) Chemoproteomics-based kinome profiling and target deconvolution of clinical multi-kinase inhibitors in primary chronic lymphocytic leukemia cells, Leukemia 25, 89-100. 10. Sumi, N. J., Kuenzi, B. M., Knezevic, C. E., Remsing Rix, L. L., and Rix, U. (2015) Chemoproteomics Reveals Novel Protein and Lipid Kinase Targets of Clinical CDK4/6 Inhibitors in Lung Cancer, ACS Chem Biol 10, 2680-2686. 11. Lanning, B. R., Whitby, L. R., Dix, M. M., Douhan, J., Gilbert, A. M., Hett, E. C., Johnson, T. O., Joslyn, C., Kath, J. C., Niessen, S., Roberts, L. R., Schnute, M. E., Wang, C., Hulce, J. J., Wei, B., Whiteley, L. O., Hayward, M. M., and Cravatt, B. F. A road map to evaluate the proteomewide selectivity of covalent kinase inhibitors, Nat Chem Biol 10, 760-767. 12. Liu, N., Hoogendoorn, S., van de Kar, B., Kaptein, A., Barf, T., Driessen, C., Filippov, D. V., van der Marel, G. A., van der Stelt, M., and Overkleeft, H. S. (2015) Direct and two-step bioorthogonal probes for Bruton's tyrosine kinase based on ibrutinib: a comparative study, Org Biomol Chem 13, 5147-5157. 13. Ranjitkar, P., Perera, B. G., Swaney, D. L., Hari, S. B., Larson, E. T., Krishnamurty, R., Merritt, E. A., Villen, J., and Maly, D. J. (2012) Affinity-based probes based on type II kinase inhibitors, J Am Chem Soc 134, 19017-19025. 14. Abo, M., and Weerapana, E. (2015) A Caged Electrophilic Probe for Global Analysis of Cysteine Reactivity in Living Cells, J Am Chem Soc 137, 7087-7090. 15. Auchinvole, C. A., Richardson, P., McGuinnes, C., Mallikarjun, V., Donaldson, K., McNab, H., and Campbell, C. J. (2012) Monitoring intracellular redox potential changes using SERS nanosensors, ACS Nano 6, 888-896. 16. Winterbourn, C. C., and Metodiewa, D. (1999) Reactivity of biologically important thiol compounds with superoxide and hydrogen peroxide, Free Radic Biol Med 27, 322-328. 17. Sadler, N. C., Melnicki, M. R., Serres, M. H., Merkley, E. D., Chrisler, W. B., Hill, E. A., Romine, M. F., Kim, S., Zink, E. M., Datta, S., Smith, R. D., Beliaev, A. S., Konopka, A., and Wright, A. T.


Page 11 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(2014) Live cell chemical profiling of temporal redox dynamics in a photoautotrophic cyanobacterium, ACS Chem Biol 9, 291-300. 18. Evans, E. K., Tester, R., Aslanian, S., Karp, R., Sheets, M., Labenski, M. T., Witowski, S. R., Lounsbury, H., Chaturvedi, P., Mazdiyasni, H., Zhu, Z., Nacht, M., Freed, M. I., Petter, R. C., Dubrovskiy, A., Singh, J., and Westlin, W. F. Inhibition of Btk with CC-292 provides early pharmacodynamic assessment of activity in mice and humans, J Pharmacol Exp Ther 346, 219-228. 19. Wu, H., Wang, W., Liu, F., Weisberg, E. L., Tian, B., Chen, Y., Li, B., Wang, A., Wang, B., Zhao, Z., McMillin, D. W., Hu, C., Li, H., Wang, J., Liang, Y., Buhrlage, S. J., Liang, J., Liu, J., Yang, G., Brown, J. R., Treon, S. P., Mitsiades, C. S., Griffin, J. D., Liu, Q., and Gray, N. S. (2014) Discovery of a potent, covalent BTK inhibitor for B-cell lymphoma, ACS Chem Biol 9, 10861091. 20. Hopf, C., Eberhard, D., Boesche, M., Bastuck, S., Dumpelfeld, B., and Bantscheff, M. (2012) Determination of kinase inhibitor potencies in cell extracts by competition binding assays and isobaric mass tags, Methods Mol Biol 803, 141-155. 21. Byrd, J. C., Harrington, B., O'Brien, S., Jones, J. A., Schuh, A., Devereux, S., Chaves, J., Wierda, W. G., Awan, F. T., Brown, J. R., Hillmen, P., Stephens, D. M., Ghia, P., Barrientos, J. C., Pagel, J. M., Woyach, J., Johnson, D., Huang, J., Wang, X., Kaptein, A., Lannutti, B. J., Covey, T., Fardis, M., McGreivy, J., Hamdy, A., Rothbaum, W., Izumi, R., Diacovo, T. G., Johnson, A. J., and Furman, R. R. (2016) Acalabrutinib (ACP-196) in Relapsed Chronic Lymphocytic Leukemia, N Engl J Med 374, 323-332. 22. Ninomiya-Tsuji, J., Kajino, T., Ono, K., Ohtomo, T., Matsumoto, M., Shiina, M., Mihara, M., Tsuchiya, M., and Matsumoto, K. (2003) A resorcylic acid lactone, 5Z-7-oxozeaenol, prevents inflammation by inhibiting the catalytic activity of TAK1 MAPK kinase kinase, J Biol Chem 278, 18485-18490. 23. Sorrell, F. J., Szklarz, M., Abdul Azeez, K. R., Elkins, J. M., and Knapp, S. (2016) Family-wide Structural Analysis of Human Numb-Associated Protein Kinases, Structure 24, 401-411. 24. Wu, J., Zhang, M., and Liu, D. (2016) Acalabrutinib (ACP-196): a selective second-generation BTK inhibitor, J Hematol Oncol 9, 21. 25. Mathea, S., Abdul Azeez, K. R., Salah, E., Tallant, C., Wolfreys, F., Konietzny, R., Fischer, R., Lou, H. J., Brennan, P. E., Schnapp, G., Pautsch, A., Kessler, B. M., Turk, B. E., and Knapp, S. (2016) Structure of the Human Protein Kinase ZAK in Complex with Vemurafenib, ACS Chem Biol 11, 1595-1602. 26. Marcotte, D. J., Liu, Y. T., Arduini, R. M., Hession, C. A., Miatkowski, K., Wildes, C. P., Cullen, P. F., Hong, V., Hopkins, B. T., Mertsching, E., Jenkins, T. J., Romanowski, M. J., Baker, D. P., and Silvian, L. F. (2010) Structures of human Bruton's tyrosine kinase in active and inactive conformations suggest a mechanism of activation for TEC family kinases, Protein Sci 19, 429439. 27. Kim, E., Hurtz, C., Koehrer, S., Wang, Z., Balasubramanian, S., Chang, B. Y., Muschen, M., Davis, R. E., and Burger, J. A. (2017) Ibrutinib inhibits pre-BCR+ B-cell acute lymphoblastic leukemia progression by targeting BTK and BLK, Blood 129, 1155-1165.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

80x40mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 12 of 15

Page 13 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


194x260mm (300 x 300 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

250x189mm (300 x 300 DPI)


Page 14 of 15

Page 15 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


274x180mm (300 x 300 DPI)


Differential Kinobeads Profiling for Target ... - ACS Publications

Recommend Documents