New Affinity Probe Targeting VEGF Receptors for ... - ACS Publications

Article pubs.acs.org/jpr

New Affinity Probe Targeting VEGF Receptors for Kinase Inhibitor Selectivity Profiling by Chemical Proteomics Xin Ku,† Stephanie Heinzlmeir,†,‡,§ Dominic Helm,† Guillaume Médard,† and Bernhard Kuster*,†,‡,§,∥ †

Chair for Proteomics and Bioanalytics, Technische Universität München, Emil Erlenmeyer Forum 5, 85354 Freising, Germany German Cancer Consortium (DKTK), Heidelberg, Germany § German Cancer Research Center (DKFZ), Heidelberg, Germany ∥ Center for Integrated Protein Science Munich, Emil Erlenmeyer Forum 5, 85354 Freising, Germany ‡

S Supporting Information *

ABSTRACT: Solid tumors are dependent for growth on nutrients and the supply of oxygen, which they often acquire via neoangiogenesis. Vascular endothelial growth factors and the corresponding receptors (VEGFRs) play central roles in this process, and consequently, the blockade of this pathway is one therapeutic strategy for cancer treatment. A number of small molecules inhibiting VEGFR inhibitors have been developed for clinical use, and a comprehensive view of target selectivity is important to assess the therapeutic as well as risk potential of a drug molecule. Recent advances in mass spectrometry-based chemical proteomics allow analyses of drug− target interactions under close-to-physiological conditions, and in this study, we report on the design, synthesis, and application of a small molecule affinity probe as a tool for the selectivity profiling of VEGFR and other kinase inhibitors. The probe is capable of binding >132 protein kinases, including angiokinases such as VEGFRs, PDGFRs, and c-KIT from lysates of cancer cell lines or human placenta tissue. Combining the new probe with Kinobeads in competitive binding assays, we were able to identify nanomolar off-targets of the VEGFR/PDGFR inhibitors pazopanib and axitinib. Because of its broad binding spectrum, the developed chemical tool can be generically used for the discovery of kinase inhibitor targets, which may contribute to a more comprehensive understanding of the mechanisms of action of such drugs. KEYWORDS: chemical proteomics, kinobeads, pazopanib, axitinib

■

INTRODUCTION Protein kinases represent an important target class for cancer therapy because of their pivotal roles in tumorigenesis and metastasis.1 As a result, considerable efforts have been devoted to the development of small molecules targeting kinases, and hundreds of active molecules are undergoing clinical trials.2 Because of the high structural conservation of the ATP-binding site of the more than 500 human protein kinases, many drugs used clinically or under development have been found to be more promiscuous than originally expected. The observed polypharmacology may lead to adverse side effects and but may also contribute to clinical efficacy. For example, Imatinib was first discovered as a potent BCR-ABL inhibitor for the treatment of chronic myelogenous leukemia (CML), and it was later found to be active against c-KIT and PDGFR, which resulted in the extension of the applications of the drug to the treatment of gastrointestinal tumors (GISTs) and further myeloproliferative diseases.3 Therefore, elucidating the complete target interaction spectrum of a drug molecule is an important step toward a full understanding of its mode of action (MoA).4−6 Conventional selectivity assessments for kinase inhibitors typically rely on biochemical assays using © 2014 American Chemical Society

purified recombinant proteins or catalytically active fragments thereof. Because of the absence of, for example, regulatory domains, cellular interaction partners, or appropriate posttranslational modifications (PTMs), the recombinant protein may not necessarily represent the relevant conformation and activity of an enzyme in its physiological context.7−9 As a consequence, binding or activity inhibition data generated from such assays may lead to difficulties in correlating in vitro assay results with the in vivo efficacy of a molecule. Developments in chemical proteomics, notably the use of immobilized small molecule inhibitors or active site ligands as affinity probes together with a quantitative mass spectrometrybased protein readouts, have enabled a more generic and unbiased approach to drug target−interaction analysis.8,10−14 There is growing evidence in the literature for the effectiveness of this approach for a number of drug target classes.10,15−18 Kinobeads, a collection of complementary nonselective immobilized kinase inhibitors, exemplify one such technology that, when configured as a competition binding assay, allows Received: December 16, 2013 Published: April 9, 2014 2445

dx.doi.org/10.1021/pr401247t | J. Proteome Res. 2014, 13, 2445−2452

Journal of Proteome Research

Article

Figure 1. Molecular docking study of BIBF1120 based on the crystal structure of VEGFR2 (PDB 3C7Q). (A) Two-dimensional interaction map of BIBF1120 in the ATP binding pocket of VEGFR2. (B) Binding pose of Compound 18 in the ATP binding pocket of VEGFR2. (C) 3D binding poses of Compound 18 (dark blue surface) in the VEGFR2 binding pocket (gray mesh), showing that the introduced linker extends into the solvent as required for subsequent immobilization of the probe.

tissues (Supplemental Figure S2 in the Supporting Information). Analyses of the available cocrystal structures and structure activity relationships (SARs) of these molecules lead to the prioritization of the phase III clinical compound BIBF1120 (Nintedanib), a 6-methoxycarbonyl-substituted indolinone derivative, reported to have high inhibitory activity against multiple angiokinases (VEGFRs, PDGFRs, and FGFRs, IC50 20−100 nM).27 The cocrystal structure of BIBF1120 bound to the intracellular kinase domain of VEGFR2 (Figure 1A) reveals the binding mode of the inhibitor at the atomic level28 and provides a rationale for the placement of a linker for later immobilization. BIBF1120 forms two hydrogen bonds with the enzyme backbone on Cys919 and Glu917 in the hinge region. The only solvent-exposed part is the methyl piperazinyl group with the 4-nitrogen atom of the N-methyl piperazinyl moiety tethered by a bidentate ionic interaction with the carboxylate oxygens of Glu850. The fixed orientation of the methyl piperazinyl group toward the solvent strongly suggests that this position would accommodate a linker for immobilization without impairing target binding. Multiple sequence alignments (Supplemental Figure S3 in the Supporting Information) show that this Glu residue is conserved in VEGFR3, PDGFR alpha and beta, as well as c-KIT, indicating that linkage at this position would be tolerable for binding other angiokinases too. On the basis of this hypothesis, a probe was designed that would maintain the salt bridge to Glu850 and carrying a primary amine-terminated PEG linker for immobilization to beads (designated as Compound 18, Figure 1B).

label-free and quantitative interaction analysis of a compound with hundreds of endogenous human kinases and other proteins as expressed in the target tissue of interest.8,10,16,17,19−26 One important design principle for affinity tools focused on a target class is that the ligands should be selective for the target class but unselective within the class. Given that such an idealized molecule has not yet been identified, we and others10,21,23,26 have been actively working on improving the coverage of the kinome by developing additional chemical probes with a view to filling important existing gaps. More specifically, we have previously reported on broadly selective probes targeting AKTs22 and FGFRs,20 both of which were not well represented by the original version of Kinobeads. In this study, we report on a probe targeting VEGFRs as yet another example of this fruitful strategy and illustrate its application by the selectivity profiling of the clinical VEGFR inhibitors axitinib and pazopanib.

■

RESULTS AND DISCUSSION

Probe Design and Synthesis

In keeping with the idea that immobilized inhibitors addressing a drug target class should be potent but unselective, we surveyed the literature and selected nine VEGFR inhibitors representing eight different scaffolds (Supplemental Figure S1 in the Supporting Information) as a starting point for probe design aiming to complement the current Kinobeads probe set that is not very efficient in capturing VEGFRs from cells and 2446



Article

Figure 2. Characterization of the kinase binding profile of Compound 18. (A) Mapping of all kinases identified in pulldowns from placenta (red) and mixed cell line lysates (blue) on the phylogenetic kinome tree. Illustration reproduced courtesy of Cell Signaling Technology. (B) Comparison of duplicates of Compound 18 (CPD18) and Kinobeads (KB) pulldowns for the enrichment of kinases from lysates of a four cell line mix (CM) and human placenta (Pla). The heat map coloring uses the log10-transformed label-free quantification (LFQ) values from the MaxQuant for protein quantification.

Molecular docking analysis confirmed that the designed probe binds in the same way to VEGFR2 as does BIBF1120 with all key interactions preserved (Figure 1B). The docking data also suggested that, as expected, the linker moiety reaches into solvent space without interrupting the binding pose of BIBF1120 (Figure 1C). Compound 18 was prepared following the synthetic scheme described by Roth et al.,27 replacing methylpiperazine by Boc-piperazine as the solvent pointing moiety. The linker was then attached to the deprotected piperazine to obtain the probe in a total of 9 steps. (See the Supporting Information for details.)

placenta lysate and 116 protein kinases from the cell mix. (See Supplemental Figure S4 and Supplemental Table S1 in the Supporting Information.) Among these are VEGFRs, PDGFRs, and KIT, the main targets of the lead compound BIBF1120, which meet a primary objective of the study. In addition, visualization of the identified kinases on the phylogenetic kinome tree (Figure 2A) revealed a broad representation of all major kinase families, thus meeting a second important objective of the probe design. We next compared the kinase binding profile of Compound 18 to that of Kinobeads (Supplemental Figure S4 in the Supporting Information). Out of the 105 protein kinases enriched by the probe from placenta lysate, 12 were more efficiently captured by Compound 18 compared with Kinobeads (using a two-fold higher average MS intensity as a criterion). Similarly, Compound 18 also led to an improvement over Kinobeads for 12 kinases using the cell mix (Figure 2B and Supplemental Table S1 in the Supporting Information). Importantly, VEGFRs, PDGFRs, and KIT were captured 10- to >100-fold more efficient with Compound 18 than using Kinobeads, depending on whether placenta or cell mix lysates were used. This difference is likely due to strong abundance differences of these targets in the respective lysates and the comparatively weak affinity of Kinobeads toward these proteins. In addition, PTK7, RIPK1, and STK24 were exclusively captured by Compound 18, demonstrating the overall success of the probe design. Multiple sequence alignment of the 16 kinases preferentially enriched by Compound 18 show that VEGFR2/ 3, PDGFRA/B, and KIT all contain the conserved Glu residue previously mentioned, providing a rational to why they are so efficiently captured. The other 11 kinases do not contain this sequence feature, suggesting that the presence of the Glu

Characterization of the Protein Binding Profiles of Immobilized Compound 18

With the probe in hand, we first determined its overall protein binding profile in mixed lysates of four cancer cell lines (K562, COLO205, OVCAR8, and SKNBE2, hereinafter referred to as cell mix) and human placenta tissue lysate (all experiments in duplicate). In total, the pulldown experiments identified 931 proteins including 132 protein kinases, making Compound 18 one of most promiscuous kinase affinity probes described to date (Figure 2A and Supplemental Table S1 in the Supporting Information). Although the number of kinases account for only 14% of the total number of captured proteins, they account for 40% of the protein amount present on the beads (judged by the mass spectrometric signal intensity). Gene ontology (GO) analysis showed that 33% of the identified proteins were classified as nucleotide binding proteins in placenta and cell mix (Supplemental Table S2 in the Supporting Information), which is not surprising given that BIBF1120 as the molecule underlying the probe design is an ATP competitive drug. Compound 18 captured 105 protein kinases from human 2447



Article

Figure 3. Results of chemical proteomic drug competition assays. (A) Chemical structures of axitinib and pazopanib. (B) Histogram dissociation constants (Kd) obtained for axitinib and pazopanib of target proteins. (C) Dose-response curves of targets of pazopanib. (D) Same as panel C but for axitinib.

reduction of bead binding, while the protein amounts of nontarget proteins remain unaffected. Binding (or loss thereof) can be quantified by intensity-based label-free LC−MS/MS analysis and half-maximal inhibitory concentration (IC50) of bead binding as well as binding constants (Kd) can be derived from the dose-response curves, as described.16,20,31 Comparison of axitinib and pazopanib showed clear similarities as well as differences in their respective selectivity profile. As expected, both drugs show potent inhibition of the primary target VEGFR3, which is in good agreement with the published data using recombinant kinase assays (Table 1).5,6,32 Bead binding of PDGFRβ was also diminished by pazopanib and axitinib in a dose-dependent manner but with lower potency than expected from the recombinant assay data.5,6,32,33 This observation may be rationalized by a number of factors including (i) the fact that PDGFRβ is generally autoinhibited in cells to keep the receptor inactive in the absence of ligand17 (as a result, pazopanib and axitinib may not bind this conformation very effectively (while the immobilized compounds can)) and (ii) the fact that in vitro kinase assays utilize isolated recombinant kinase (or just the kinase domain) and thus lack auxiliary proteins, cofactors, or naturally occurring post-translational modifications that may impact on drug−protein interactions.

residue is advantageous but not essential for probe binding (Figure S3 in the Supporting Information). Selectivity Profiling of Axitinib and Pazopanib

We next applied Compound 18 in conjunction with the established set of affinity probes used in the Kinobeads approach to determine the selectivity profiles of two clinical VEGFR inhibitors by chemical proteomics (Figure 3A, see Methods section and Supplemental Tables S3 and S4 in the Supporting Information). Axitinib is a pan-VEGFR inhibitor approved in 2012 for the treatment of advanced renal cell carcinoma, and pazopanib is a multitarget kinase inhibitor affecting several angiokinases and was approved in 2009 for the treatment of soft tissue sarcoma as well as renal cell carcinoma.29,30 We chose the cell mix (see previous) for this experiment to expose the drugs to a broad range of natively expressed kinases and other nucleotide binding proteins, acknowledging that the cell mix is not the ideal protein source of VEGFRs and other angiogenesis-related proteins. Pulldowns using Compound 18 plus Kinobeads were performed in a competition setup using lysates preincubated with six different concentrations (0, 2.5, 25, 250, 2500, and 25 000 nM) of pazopanib or axitinib, respectively. (For a schematic workflow, see Supplemental Figure S5 in the Supporting Information.) In this assay, targets of any one drug exhibit a dose-dependent 2448



Article

observed discrepancy between the chemical proteomic and in vitro kinase assay data for AURKA is a general feature of AURKA inhibitors, we performed additional chemical proteomics experiments for four further AURK inhibitors (Figure S6 in the Supporting Information). For AMG900 and MLN8054, the in vitro kinase assay and chemical proteomic data are in excellent agreement. For SNS314, the in vitro kinase assay potency is 35 times higher, but for GSK1070916, it is six times lower than determined by chemical proteomics. Hence, the five AURK inhibitors investigated do not form a coherent picture in this regard, which currently precludes any conclusive answer to the previous question. More generally speaking, we and others have observed that IC50 values determined by chemical proteomics are most of the time higher than those measured by in vitro kinase assays. The assay conditions are obviously very different and, for example, the high concentration of cellular protein in the chemical proteomic assay will lead to considerable drug absorption, which in turn lowers the concentration of the free drug available for exerting its inhibitory effect. In addition, the chemical proteomic assay uses the ATP concentration of the lysate, while in vitro kinase assays are often performed with Km-adjusted ATP concentrations. The aforementioned presence/absence of other cellular factors will likely also influence assay results, and ordinary experimental mistakes in large-scale experiments may also lead to discrepancies.

Table 1. Inhibitory Concentrations and Binding Constants of Pazopanib and Axitinitib for Protein Targets As Determined in this Study (KB assay) Compared with Potency Values Determined by Recombinant Kinase Assays Reported in the Literature33 pazopanib

target

KB assay IC50

KB assay Kd

ABL1 ABL2 AURKA AURKB DDR1 DDR2 LIMK1 MAP4K2 PDGFRβ RET RIPK2 TNIK VEGFR3

2.78 >10 2.97 >10 0.26 0.32 1.42 2.84 0.67 0.26 2.61 1.38 0.02

1.61 >10 0.65 >10 0.14 0.23 0.70 1.16 0.63 0.11 0.78 1.38 0.01

axitinib recomb. assay, Kd 0.62 3 7.1/0.17a 0.057 0.098 0.72 2.7 0.002 0.31 0.58 0.31 0.027

KB assay IC50

KB assay Kd

2.61 2.51 2.34 1.25 3.30 >10 2.32 3.08 0.62 0.25 >10 3.38 0.007

1.07 0.53 0.37 0.83 1.42 >10 0.49 1.45 0.59 0.08 >10 2.40 0.003

recomb. assay, Kd 0.084 0.07 0.072 0.011 0.34 5.3 1.3 0.006 0.12 9.9 0.18 0.17

a

IC50 value obtained by in vitro kinase activity assay using recombinant full length AURKA.42

■

Apart from the primary targets, both drugs show medium to high interaction potency to several other kinases (Figures 3 and Table 1). Pazopanib showed submicromolar affinity to six further kinases in our chemical proteomic assay: AURKA (∼80fold window over VEGFR3), DDR1/2 (∼20-fold window), LIMK1 (∼80-fold window), RET (∼10-fold window), and RIPK2 (∼100-fold window). For axitinib, five such cases were identified: ABL2 (∼180-fold window), AURKA/B (∼10-fold window, LIMK1 (∼160-fold window), and RET (∼25 fold window). Depending on the pharmacokinetic and pharmacodynamic properties of the inhibitors, the previous potencies may be of physiological relevance. The AURKA/B and RET oncogenes are often overexpressed in tumor cells,34−38 both are reported targets of axitinib in vitro,5,6,32 and, according to our data, the drug also appears to be active against these targets in cell lysates. In recombinant kinase assays, axitinib showed no binding affinity to LIMK1 at a concentration of 10 μM.33 It was, however, detected in our assay with a Kd of 490 nM, suggesting that this interaction may have been missed in previous studies. LIMK1 is a critical regulator of actin dynamics39 and has been shown to play a regulatory role in tumor cell invasion40 and is also involved in tumor cell-induced angiogenesis.41 Therefore, simultaneously targeting AURKA/B, RET, and LIMK1 may provide additional therapeutic benefit for the treatment of cancer using axitinib. Reported off-targets of pazopanib include ABL1, RET, DDR1, DDR2, LIMK1, RIPK2, as well as TNIK,5,6,32 and these targets were broadly recapitulated in our assay (Table 1). Interestingly, we initially observed a strong discrepancy in the data for AURKA; while the reported in vitro kinase assay data using kinase domain for pazopanib determined a Kd of 7.1 μM, our binding data showed a Kd of 650 nM. The latter is more in line with a recent study by Isham et al.,42 who reported an inhibitory concentration of 167 nM when using the full length protein. The same authors showed that pazopanib synergizes with paclitaxel in anaplastic thyroid cancer lines and xenografts, providing strong evidence that the drug also inhibits AURKA in vivo. To investigate if the

CONCLUSIONS In summary, we have designed and synthesized a small molecular probe that enables the affinity enrichment of more than 130 protein kinases from lysates of human cancer cell lines and tissues. As intended, the probe efficiently binds angiokinases such as VEGFRs, PDGFRs, and KIT as well as a few other kinases not covered by the established Kinobeads approach. The probe therefore is a useful addition to the chemical tool box for, for example, profiling the expression of kinases in human cells or the determining the selectivity of kinase inhibitors, and the results obtained validate our structure-based design approach for the development of broadly selective affinity tools. Compound 18 exemplifies our ongoing efforts to eventually make all human kinases accessible to chemical proteomic experiments. We and others have been trying to identify a “one compound Kinobead” and some progress has been made along this line.10,21,23,26 However, no genuine pan-kinase probe has been identified yet, and, as things stand, an all-encompassing probe appears out of reach. As a result, our strategy is and has been to fill important gaps in kinome coverage instead. While it is possible to increase kinome coverage in this way, mixing increasing numbers of immobilized probes also impose some limitations (e.g., diluting out some of the protein binders). One way of addressing this in the future is to generate “thematic” Kinobeads by mixing a (select few) number of probes in a way that is tailored toward a desired kinase family, say angiogenic targets exemplified by Compound 18.

■

MATERIALS AND METHODS A more detailed description of the materials and methods can be found in the Supporting Information. Molecular Docking

The crystal structure of VEGFR2 in complex with BIB1120 was retrieved from the Protein Data Bank (PDB code 3C7Q), and 2449



Article

the binding pose of Compound 18 as well as other molecules in the ATP-binding site of VEGFR2 was predicted by the software Glide (Schrödinger).

values were calculated using a depletion factor, as described.16,20,31

Chemical Synthesis

All raw mass spectrometry files are available from https://www. proteomicsdb.org/proteomicsdb/#projects/4160.

Data Availability

All experimental details and structural characterizations of the probe (Compound 18) and intermediates can be found in the Supporting Information. The final product was obtained with >95% purity. Kinobeads compounds were purchased or synthesized as described,10 and the structures of the affinity probes used in this study are shown in the Supporting Information. We note that the set of probes referred to as Kinobeads in this study deviates from the one originally published by Bantscheff et al.10

■

ASSOCIATED CONTENT

S Supporting Information *

Detailed materials and methods, chemical structures of selected VEGFR kinase inhibitors from the literature considered as potential starting points to design VEFGR affinity probe, sequence alignment of 16 kinases that were preferentially enriched by Compound 18 composition of Kinobeads used in this study, comparison of proteins and kinases captured by Compound 18 and Kinobeads in a mixed lysate of four human cancer cell lines and in placenta lysate. Schematic representation of the workflow used for drug selectivity profiling. Doseresponse curves for AURKA of four aurora kinase inhibitors determined by Kinobeads competition assay. Intensity (LFQ, label free quantification). Gene Ontology analysis of the total protein identified by CPD18. Replicates for selectivity profiling of axitinib and pazopanib. This material is available free of charge via the Internet at http://pubs.acs.org.

Human Placenta Tissue and Cancer Cell Lines

Postdelivery human placenta tissue was obtained from Freising hospital following informed consent by the donor. OVCAR8, K562, COLO205, and SKNBE2 cells were cultivated using standard conditions. Tissue and cells were lysed in the presence of protease and phosphatase inhibitors. Homogenates were clarified by ultracentrifugation and stored at −80 °C until further use. Compound Immobilization

■

Compound 18 was immobilized on NHS-activated sepharose beads at a coupling density of 2 μmol of the probe per 1 mL of settled beads, as described.10,20,22 The functionalized beads were stored in ethanol at 4 °C in the dark until use. For drugprofiling experiments, Compound 18 beads were mixed in equal amounts with Kinobeads probes.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel. +49 8161 715696. Fax: +49 8161 715931.

Pulldowns and Competition Assay

Notes

Pulldowns using Compound 18 beads were performed in duplicate using 5 mg of protein for each experiment. Drug competition assays were performed similarly (in triplicate) and as described previously.10,16,20,22,24 In brief, lysates were incubated with the respective drug in six concentrations (DMSO, 2.5 nM, 25 nM, 250 nM, 2.5 μM, 25 μM), followed by incubation with mixed beads (Kinobeads plus Compound 18 beads, 100 μL total settled amount). Bound proteins were eluted with LDS sample buffer, reduced and alkylated, and run into a 4−12% NuPAGE gel for ∼1 cm. In-gel trypsin digestion was performed according to standard procedures.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dr. Barbara Suess and Prof. Dr. Thomas Hofmann of the Chair of Food Chemistry and Molecular Sensory Science, Technische Universität München for NMR analysis. We thank Dr. Sabine Schweizer, Prof. Dr. Iris Antes at the Technische Universität München, and Dr. Xiaofeng Liu at the East China University of Science and Technology for assistance with docking studies. We also thank Benjamin Ruprecht for MS measurements and Andreas Klaus, Michaela Kroetz-Fahning, and Andrea Hubauer for technical assistance.

Liquid Chromatography Tandem Mass Spectrometry (LC−MS/MS) Analysis

■

Peptides generated by in-gel trypsin digestion were analyzed by LC−MS/MS analysis on a nanoLC-Ultra 1D+ (Eksigent) coupled to a LTQ-Orbitrap Velos mass spectrometer (ThermoFisher Scientific) using a 210 min acetonitrile gradient and subjecting up to 10 peptide precursors to fragmentation by higher energy collision-induced dissociation (HCD).

REFERENCES

(1) Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.; Sudarsanam, S. The protein kinase complement of the human genome. Science 2002, 298 (5600), 1912−1934. (2) Cohen, P. Protein kinases–the major drug targets of the twentyfirst century? Nat. Rev. Drug Discovery 2002, 1 (4), 309−315. (3) Drews, J. Case histories, magic bullets and the state of drug discovery. Nat. Rev. Drug Discovery 2006, 5 (8), 635−640. (4) Knapp, S.; Arruda, P.; Blagg, J.; Burley, S.; Drewry, D. H.; Edwards, A.; Fabbro, D.; Gillespie, P.; Gray, N. S.; Kuster, B.; Lackey, K. E.; Mazzafera, P.; Tomkinson, N. C.; Willson, T. M.; Workman, P.; Zuercher, W. J. A public-private partnership to unlock the untargeted kinome. Nat. Chem. Biol. 2012, 9 (1), 3−6. (5) Karaman, M. W.; Herrgard, S.; Treiber, D. K.; Gallant, P.; Atteridge, C. E.; Campbell, B. T.; Chan, K. W.; Ciceri, P.; Davis, M. I.; Edeen, P. T.; Faraoni, R.; Floyd, M.; Hunt, J. P.; Lockhart, D. J.; Milanov, Z. V.; Morrison, M. J.; Pallares, G.; Patel, H. K.; Pritchard, S.; Wodicka, L. M.; Zarrinkar, P. P. A quantitative analysis of kinase inhibitor selectivity. Nat. Biotechnol. 2008, 26 (1), 127−132.

Peptide and Protein Identification and Quantification

The raw MS spectra were processed using MaxQuant (version 1.4.0.5), and MS/MS spectra were searched against the IPI human database (version 3.68) using Andromeda. Search results were filtered for 1% peptide and protein FDR. Data Analysis

Proteins that displayed a dose-dependent inhibition of bead binding were selected and analyzed in GraphPad Prism. Halfmaximal inhibition of binding concentrations (IC50) was calculated by nonlinear regression with variable slope and the constraint of the DMSO control value to be equal to 1. Kd 2450



Article

profiling using chemoproteomics. Methods Mol. Biol. 2012, 795, 161− 177. (24) Wu, Z.; Doondeea, J. B.; Gholami, A. M.; Janning, M. C.; Lemeer, S.; Kramer, K.; Eccles, S. A.; Gollin, S. M.; Grenman, R.; Walch, A.; Feller, S. M.; Kuster, B. Quantitative chemical proteomics reveals new potential drug targets in head and neck cancer. Mol. Cell. Proteomics 2011, 10 (12), M111 011635. (25) Wu, Z.; Moghaddas Gholami, A.; Kuster, B. Systematic identification of the HSP90 candidate regulated proteome. Mol. Cell. Proteomics 2012, 11 (6), M111 016675. (26) Zhang, L.; Holmes, I. P.; Hochgrafe, F.; Walker, S. R.; Ali, N. A.; Humphrey, E. S.; Wu, J.; de Silva, M.; Kersten, W. J.; Connor, T.; Falk, H.; Allan, L.; Street, I. P.; Bentley, J. D.; Pilling, P. A.; Monahan, B. J.; Peat, T. S.; Daly, R. J. Characterization of the novel broad-spectrum kinase inhibitor CTx-0294885 as an affinity reagent for mass spectrometry-based kinome profiling. J. Proteome Res. 2013, 12 (7), 3104−3116. (27) Roth, G. J.; Heckel, A.; Colbatzky, F.; Handschuh, S.; Kley, J.; Lehmann-Lintz, T.; Lotz, R.; Tontsch-Grunt, U.; Walter, R.; Hilberg, F. Design, synthesis, and evaluation of indolinones as triple angiokinase inhibitors and the discovery of a highly specific 6methoxycarbonyl-substituted indolinone (BIBF 1120). J. Med. Chem. 2009, 52 (14), 4466−4480. (28) Hilberg, F.; Roth, G. J.; Krssak, M.; Kautschitsch, S.; Sommergruber, W.; Tontsch-Grunt, U.; Garin-Chesa, P.; Bader, G.; Zoephel, A.; Quant, J.; Heckel, A.; Rettig, W. J. BIBF 1120: triple angiokinase inhibitor with sustained receptor blockade and good antitumor efficacy. Cancer Res. 2008, 68 (12), 4774−4782. (29) Deeks, E. D. Pazopanib: in advanced soft tissue sarcoma. Drugs 2012, 72 (16), 2129−2140. (30) Hu-Lowe, D. D.; Zou, H. Y.; Grazzini, M. L.; Hallin, M. E.; Wickman, G. R.; Amundson, K.; Chen, J. H.; Rewolinski, D. A.; Yamazaki, S.; Wu, E. Y.; McTigue, M. A.; Murray, B. W.; Kania, R. S.; O’Connor, P.; Shalinsky, D. R.; Bender, S. L. Nonclinical antiangiogenesis and antitumor activities of axitinib (AG-013736), an oral, potent, and selective inhibitor of vascular endothelial growth factor receptor tyrosine kinases 1, 2, 3. Clin. Cancer Res. 2008, 14 (22), 7272−7283. (31) Sharma, K.; Weber, C.; Bairlein, M.; Greff, Z.; Keri, G.; Cox, J.; Olsen, J. V.; Daub, H. Proteomics strategy for quantitative protein interaction profiling in cell extracts. Nat. Methods 2009, 6 (10), 741− 744. (32) Sonpavde, G.; Hutson, T. E.; Rini, B. I. Axitinib for renal cell carcinoma. Expert Opin. Invest. Drugs 2008, 17 (5), 741−748. (33) Davis, M. I.; Hunt, J. P.; Herrgard, S.; Ciceri, P.; Wodicka, L. M.; Pallares, G.; Hocker, M.; Treiber, D. K.; Zarrinkar, P. P. Comprehensive analysis of kinase inhibitor selectivity. Nat. Biotechnol. 2011, 29 (11), 1046−1051. (34) Takaya, K.; Yoshimasa, T.; Arai, H.; Tamura, N.; Miyamoto, Y.; Itoh, H.; Nakao, K. Expression of the RET proto-oncogene in normal human tissues, pheochromocytomas, and other tumors of neural crest origin. J. Mol. Med. 1996, 74 (10), 617−621. (35) Marumoto, T.; Zhang, D.; Saya, H. Aurora-A - a guardian of poles. Nat. Rev. Cancer 2005, 5 (1), 42−50. (36) Fu, J.; Bian, M.; Jiang, Q.; Zhang, C. Roles of Aurora kinases in mitosis and tumorigenesis. Mol. Cancer Res. 2007, 5 (1), 1−10. (37) Borrello, M. G.; Ardini, E.; Locati, L. D.; Greco, A.; Licitra, L.; Pierotti, M. A. RET inhibition: implications in cancer therapy. Expert Opin. Ther. Targets 2013, 17 (4), 403−419. (38) Mehra, R.; Serebriiskii, I. G.; Burtness, B.; Astsaturov, I.; Golemis, E. A. Aurora kinases in head and neck cancer. Lancet Oncol. 2013, 14 (10), e425−e435. (39) Khurana, T.; Khurana, B.; Noegel, A. A. LIM proteins: association with the actin cytoskeleton. Protoplasma 2002, 219 (1−2), 1−12. (40) Yoshioka, K.; Foletta, V.; Bernard, O.; Itoh, K. A role for LIM kinase in cancer invasion. Proc. Natl. Acad. Sci. U. S. A. 2003, 100 (12), 7247−7252.

(6) Anastassiadis, T.; Deacon, S. W.; Devarajan, K.; Ma, H.; Peterson, J. R. Comprehensive assay of kinase catalytic activity reveals features of kinase inhibitor selectivity. Nat. Biotechnol. 2011, 29 (11), 1039−1045. (7) Zinn, N.; Hopf, C.; Drewes, G.; Bantscheff, M. Mass spectrometry approaches to monitor protein-drug interactions. Methods 2012, 57 (4), 430−440. (8) Schirle, M.; Bantscheff, M.; Kuster, B. Mass spectrometry-based proteomics in preclinical drug discovery. Chem. Biol. 2012, 19 (1), 72− 84. (9) Bantscheff, M.; Drewes, G. Chemoproteomic approaches to drug target identification and drug profiling. Bioorg. Med. Chem. 2012, 20 (6), 1973−1978. (10) 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.; Drewes, G. Quantitative chemical proteomics reveals mechanisms of action of clinical ABL kinase inhibitors. Nat. Biotechnol. 2007, 25 (9), 1035− 1044. (11) Bantscheff, M.; Kuster, B. Quantitative mass spectrometry in proteomics. Anal Bioanal Chem. 2012, 404 (4), 937−938. (12) Godl, K.; Wissing, J.; Kurtenbach, A.; Habenberger, P.; Blencke, S.; Gutbrod, H.; Salassidis, K.; Stein-Gerlach, M.; Missio, A.; Cotten, M.; Daub, H. An efficient proteomics method to identify the cellular targets of protein kinase inhibitors. Proc. Natl. Acad. Sci. U. S. A. 2003, 100 (26), 15434−15439. (13) Graves, P. R.; Kwiek, J. J.; Fadden, P.; Ray, R.; Hardeman, K.; Coley, A. M.; Foley, M.; Haystead, T. A. Discovery of novel targets of quinoline drugs in the human purine binding proteome. Mol. Pharmacol. 2002, 62 (6), 1364−1372. (14) Patricelli, M. P.; Szardenings, A. K.; Liyanage, M.; Nomanbhoy, T. K.; Wu, M.; Weissig, H.; Aban, A.; Chun, D.; Tanner, S.; Kozarich, J. W. Functional interrogation of the kinome using nucleotide acyl phosphates. Biochemistry 2007, 46 (2), 350−358. (15) Liu, Y.; Patricelli, M. P.; Cravatt, B. F. Activity-based protein profiling: the serine hydrolases. Proc. Natl. Acad. Sci. U. S. A. 1999, 96 (26), 14694−14699. (16) Lemeer, S.; Zorgiebel, C.; Ruprecht, B.; Kohl, K.; Kuster, B. Comparing Immobilized Kinase Inhibitors and Covalent ATP Probes for Proteomic Profiling of Kinase Expression and Drug Selectivity. J. Proteome Res. 2013, 12 (4), 1723−1731. (17) Bantscheff, M.; Hopf, C.; Savitski, M. M.; Dittmann, A.; Grandi, P.; Michon, A. M.; Schlegl, J.; Abraham, Y.; Becher, I.; Bergamini, G.; Boesche, M.; Delling, M.; Dumpelfeld, B.; Eberhard, D.; Huthmacher, C.; Mathieson, T.; Poeckel, D.; Reader, V.; Strunk, K.; Sweetman, G.; Kruse, U.; Neubauer, G.; Ramsden, N. G.; Drewes, G. Chemoproteomics profiling of HDAC inhibitors reveals selective targeting of HDAC complexes. Nat. Biotechnol. 2011, 29 (3), 255−265. (18) Dadvar, P.; O’Flaherty, M.; Scholten, A.; Rumpel, K.; Heck, A. J. A chemical proteomics based enrichment technique targeting the interactome of the PDE5 inhibitor PF-4540124. Mol. Biosyst. 2009, 5 (5), 472−482. (19) Godl, K.; Gruss, O. J.; Eickhoff, J.; Wissing, J.; Blencke, S.; Weber, M.; Degen, H.; Brehmer, D.; Orfi, L.; Horvath, Z.; Keri, G.; Muller, S.; Cotten, M.; Ullrich, A.; Daub, H. Proteomic characterization of the angiogenesis inhibitor SU6668 reveals multiple impacts on cellular kinase signaling. Cancer Res. 2005, 65 (15), 6919−6926. (20) Ku, X.; Heinzlmeir, S.; Liu, X.; Medard, G.; Kuster, B. A new chemical probe for quantitative proteomic profiling of fibroblast growth factor receptor and its inhibitors. J. Proteomics 2014, 96 (16), 44−55. (21) Oppermann, F. S.; Gnad, F.; Olsen, J. V.; Hornberger, R.; Greff, Z.; Keri, G.; Mann, M.; Daub, H. Large-scale proteomics analysis of the human kinome. Mol. Cell. Proteomics 2009, 8 (7), 1751−1764. (22) Pachl, F.; Plattner, P.; Ruprecht, B.; Medard, G.; Sewald, N.; Kuster, B. Characterization of a chemical affinity probe targeting Akt kinases. J. Proteome Res. 2013, 12 (8), 3792−3800. (23) Schirle, M.; Petrella, E. C.; Brittain, S. M.; Schwalb, D.; Harrington, E.; Cornella-Taracido, I.; Tallarico, J. A. Kinase inhibitor 2451



Article

(41) Vlecken, D. H.; Bagowski, C. P. LIMK1 and LIMK2 are important for metastatic behavior and tumor cell-induced angiogenesis of pancreatic cancer cells. Zebrafish 2009, 6 (4), 433−439. (42) Isham, C. R.; Bossou, A. R.; Negron, V.; Fisher, K. E.; Kumar, R.; Marlow, L.; Lingle, W. L.; Smallridge, R. C.; Sherman, E. J.; Suman, V. J.; Copland, J. A.; Bible, K. C. Pazopanib enhances paclitaxelinduced mitotic catastrophe in anaplastic thyroid cancer. Sci. Transl Med. 2013, 5 (166), 166ra3.

2452


New Affinity Probe Targeting VEGF Receptors for ... - ACS Publications

Recommend Documents