Insights on Structural Characteristics and Ligand Binding Mechanisms of CDK2

Cyclin-dependent kinase 2 (CDK2) is a crucial regulator of the eukaryotic cell cycle. However it is well established that monomeric CDK2 lacks regulatory activity, which needs to be aroused by its positive regulators, cyclins E and A, or be phosphorylated on the catalytic segment. Interestingly, these activation steps bring some dynamic changes on the 3D-structure of the kinase, especially the activation segment. Until now, in the monomeric CDK2 structure, three binding sites have been reported, including the adenosine triphosphate (ATP) binding site (Site I) and two non-competitive binding sites (Site II and III). In addition, when the kinase is subjected to the cyclin binding process, the resulting structural changes give rise to a variation of the ATP binding site, thus generating an allosteric binding site (Site IV). All the four sites are demonstrated as being targeted by corresponding inhibitors, as is illustrated by the allosteric binding one which is targeted by inhibitor ANS (fluorophore 8-anilino-1-naphthalene sulfonate). In the present work, the binding mechanisms and their fluctuations during the activation process attract our attention. Therefore, we carry out corresponding studies on the structural characterization of CDK2, which are expected to facilitate the understanding of the molecular mechanisms of kinase proteins. Besides, the binding mechanisms of CDK2 with its relevant inhibitors, as well as the changes of binding mechanisms following conformational variations of CDK2, are summarized and compared. The summary of the conformational characteristics and ligand binding mechanisms of CDK2 in the present work will improve our understanding of the molecular mechanisms regulating the bioactivities of CDK2.     

Discovery of Non‐ATP‐Competitive Inhibitors of Polo‐like Kinase 1

Polo‐like kinase 1 (Plk1) is an evolutionarily conserved serine/threonine kinase, and its N‐terminal kinase domain (KD) controls cell signaling through phosphorylation. Inhibitors of Plk1 are potential anticancer drugs. Most known Plk1 KD inhibitors are ATP‐competitive compounds, which may suffer from low selectivity. In this study we discovered novel non‐ATP‐competitive Plk1 KD inhibitors by virtual screening and experimental studies. Potential binding sites in Plk1 KD were identified by using the protein binding site detection program Cavity. The identified site was subjected to molecular‐docking‐based virtual screening. The activities of top‐ranking compounds were evaluated by in vitro enzyme assay with full‐length Plk1 and direct binding assay with Plk1 KD. Several compounds showed inhibitory activity, and the most potent was found to be 3‐((2‐oxo‐2‐(thiophen‐2‐yl)ethyl)thio)‐6‐(pyridin‐3‐ylmethyl)‐1,2,4‐triazin‐5(4H)‐one (compound 4 ) with an IC₅₀ value of 13.1±1.7 μm . Our work provides new insight into the design of kinase inhibitors that target non‐ATP binding sites.     

Phospholipid synthesizing enzymes of dermatophytes: II. Characterization of choline kinase

Choline kinase was located in the cytosolic fractions of the filamentous, pathogenic fungi,Microsporum gypseum andEpidermophyton floccosum. A broad pH optima (6.0–9.0) was observed for theM. gypseum enzyme, but theE. floccosum enzyme was active at pH 8.4 as well as 10.5, the activity being higher at pH 8.4. Enzyme from both dermatophytes had Km value of 3.3×10^?4 M for choline; however, for ATP, it was 6.6×10^?4 M and 12.6×10^?4 M forM gypseum andE. floccosum, respectively. Choline kinase of both dermatophytes showed SH-group requirement. TheM. gypseum choline kinase was inhibited to a greater extent by Mn²⁺, Ca²⁺ and Ba²⁺ than was theE. floccosum enzyme. In comparison to other nucleotides, ATP was the most effective phosphate donor for phosphorylating choline in both dermatophytes. Higher concentrations of ATP inhibited the enzyme inM. gypseum as well asE. floccosum. Phosphorylcholine inhibited the choline kinase activity from both fungi, whereas phosphoethanolamine and glycerol 3-phosphate were stimulatory.     

Comparison of conservation within and between the Ser/Thr and Tyr protein kinase family: proposed model for the catalytic domain of the epidermal growth factor receptor

The protein kinase family can be subdivided into two main groupsbased on their ability to phosphorylate Ser/Thr or Tyr substrates.In order to understand the basis of this functional difference,we have carried out a comparative analysis of sequence conservationwithin and between the Ser/Thr and Tyr protein kinases. A multiplesequence alignment of 86 protein kinase sequences was generated.For each position in the alignment we have computed the conservationof residue type in the Ser/Thr, in the Tyr and in both of thekinase subfamilies. To understand the structural and/or functionalbasis for the conservation, we have mapped these conservationproperties onto the backbone of the recently determined structureof the cAMP–dependent Ser/Thr kinase. The results showthat the kinase structure can be roughly segregated, based uponconservation, into three zones. The inner zone contains residueshighly conserved in all the kinase family and describes thehydrophobic core of the enzyme together with residues essentialfor substrate and ATP binding and catalysis. The outer zonecontains residues highly variable in all kinases and representsthe solvent–exposed surface of the protein. The thirdzone is comprised of residues conserved in either the Ser/Thror Tyr kinases or in both, but which are not conserved betweenthem. These are sandwiched between the hydrophobic core andthe solvent-exposed surface. In addition to analyzing overallconservation hi the kinase family, we have also looked at conservationof its substrate and ATP binding sites. The ATP site is highlyconserved throughout the kinases, whereas the substrate bindingsite is more variable. The active site contains several positionswhich differ between the Ser/Thr and Tyr kinases and may beresponsible for discriminating between hydroxyl bearing sidechains. Using this information we propose a model for Tyr substratebinding to the catalytic domain of the epidermal growth factorreceptor (EGFR).     

ATP‐Noncompetitive Inhibitors of CDK–Cyclin Complexes

Progression through the cell division cycle is controlled by a family of cyclin‐dependent kinases (CDKs), the activity of which depends on their binding to regulatory partners (cyclins A–H). Deregulation of the activity of CDKs has been associated with the development of infectious, neurodegenerative, and proliferative diseases such as Alzheimer's, Parkinson's, or cancer. Most cancer cells contain mutations in the pathways that control the activity of CDKs. This observation led this kinase family to become a central target for the development of new drugs for cancer therapy. A range of structurally diverse molecules has been shown to inhibit the activity of CDKs through their activity as ATP antagonists. Nevertheless, the ATP binding sites on CDKs are highly conserved, limiting the kinase specificity of these inhibitors. Various genetic and crystallographic approaches have provided essential information about the mechanism of formation and activation of CDK–cyclin complexes, providing new ways to implement novel research strategies toward the discovery of new, more effective and selective drugs. Herein we review the progress made in the development of ATP‐noncompetitive CDK–cyclin inhibitors.     

Phospholipid synthesis in mammary tissue. Choline and ethanolamine kinases: Kinetic evidence for two discrete active sites

Choline and ethanolamine kinases are located in the high speed supernatant of lactating bovine mammary gland. Maximum activities of choline and ethanolamine kinases were observed at pH 9.2 and 8.0, respectively, with the rate of ethanolamine phosphorylation being 1/15 that of choline phosphorylation. Activation energies of 29 joules (Q₁₀-1.5) and 31 joules (Q₁₀=1.5) were calculated between 3.4 and 31.3C for choline kinase and ethanolamine kinase, respectively. Above 31.3 C, the Arrhenius plot deviated from linearity for both enzymes, suggesting that denaturation was occurring. An apparent Km of 0.25 mM for choline was obtained for choline kinase activity. The apparent Km of ethanolamine kinase for ethanolamine was unusually high (17 mM), and activity was not linear with increasing protein concentration. Activity was tripled and the Km decreased to 2.5 mM when the enzyme preparation was washed with butanol: benzene mixture, suggesting the presence of an endogenous competitive inhibitor(s), with respect to ethanolamine. Choline kinase was not affected by the solvent wash. Substrate competition studies revealed that choline kinase was slightly inhibited competitively by ethanolamine (apparent Ki=19–21 mM), whereas choline was a potent competitive inhibitor of ethanolamine kinase (apparent Ki=0.33–0.50 mM). The results indicated that these two kinase activities were mediated by two distinct active sites, possibly on a single protein. The significance of choline in the regulation of phosphatidylethanolamine synthesis is discussed.     

Molecular system bioenergics of the heart: experimental studies of metabolic compartmentation and energy fluxes versus computer modeling

In this review we analyze the recent important and remarkable advancements in studies of compartmentation of adenine nucleotides in muscle cells due to their binding to macromolecular complexes and cellular structures, which results in non-equilibrium steady state of the creatine kinase reaction. We discuss the problems of measuring the energy fluxes between different cellular compartments and their simulation by using different computer models. Energy flux determinations by (18)O transfer method have shown that in heart about 80% of energy is carried out of mitochondrial intermembrane space into cytoplasm by phosphocreatine fluxes generated by mitochondrial creatine kinase from adenosine triphosphate (ATP), produced by ATP Synthasome. We have applied the mathematical model of compartmentalized energy transfer for analysis of experimental data on the dependence of oxygen consumption rate on heart workload in isolated working heart reported by Williamson et al. The analysis of these data show that even at the maximal workloads and respiration rates, equal to 174 μmol O(2) per min per g dry weight, phosphocreatine flux, and not ATP, carries about 80-85% percent of energy needed out of mitochondria into the cytosol. We analyze also the reasons of failures of several computer models published in the literature to correctly describe the experimental data.     

Identification of putative binding sites of P-glycoprotein based on its homology model

A homology model of P-glycoprotein based on the crystal structure of the multidrug transporter Sav1866 is developed, incorporated into a membrane environment, and optimized. The resulting model is analyzed in relation to the functional state and potential binding sites. The comparison of modeled distances to distances reported in experimental studies between particular residues suggests that the model corresponds most closely to the first ATP hydrolysis step of the protein transport cycle. Comparison to the protein 3D structure confirms this suggestion. Using SiteID and Site Finder programs three membrane related binding regions are identified: a region at the interface between the membrane and cytosol and two regions located in the transmembrane domains. The regions contain binding pockets of different size, orientation, and amino acids. A binding pocket located inside the membrane cavity is also identified. The pockets are analyzed in relation to amino acids shown experimentally to influence the protein function. The results suggest that the protein has multiple binding sites and may bind and/or release substrates in multiple pathways.     

Directed Modulation of Protein Kinase C Isozyme Selectivity with Bisubstrate‐Based Inhibitors

Kinases present an attractive target for drug development, since they are involved in vital cellular processes and are implicated in a variety of diseases, such as cancer and diabetes. However, obtaining selectivity for a specific kinase over others is difficult since many current kinase inhibitors exclusively target the highly conserved kinase ATP binding domain. Previously, a microarray‐based strategy to discover so‐called bisubstrate‐based inhibitors that target the more specific peptide binding groove in addition to the ATP binding site was described. One attractive feature of this strategy is the opportunity to tune the selectivity of these inhibitors by systematically varying components. In an extension to this previous work, this study explores the potential of this guided selectivity modulation, leading to a series of inhibitors with different selectivity profiles against highly homologous protein kinase C (PKC) isozymes. Of the inhibitors studied, most exhibited improved potency and selectivity compared with their constituent parts. Furthermore, the selectivity was found to be tunable either through modification of the pseudosubstrate peptide (peptide binding groove) or the ATP‐competitive part (ATP binding site). In a number of cases, the selectivity of the construct could be predicted from the initial peptide substrate profiling experiment. Since this strategy is applicable to all kinase sets, it could be used to rapidly develop uniquely selective inhibitors.     

X‐ray Structural Analysis of Tau‐Tubulin Kinase 1 and Its Interactions with Small Molecular Inhibitors

Tau‐tubulin kinase 1 (TTBK1) is a serine/threonine/tyrosine kinase that putatively phosphorylates residues including S422 in tau protein. Hyperphosphorylation of tau protein is the primary cause of tau pathology and neuronal death associated with Alzheimer’s disease. A library of 12 truncation variants comprising the TTBK1 kinase domain was screened for expression in Escherichia coli and insect cells. One variant (residues 14–313) could be purified, but mass spectrometric analysis revealed extensive phosphorylation of the protein. Co‐expression with lambda phosphatase in E. coli resulted in production of homogeneous, nonphosphorylated TTBK1. Binding of ATP and several compounds to TTBK1 was characterized by surface plasmon resonance. Crystal structures of TTBK1 in the unliganded form and in complex with ATP, and two high‐affinity ATP‐competitive inhibitors, 3‐[(6,7‐dimethoxyquinazolin‐4‐yl)amino]phenol ( 1 ) and methyl 2‐bromo‐5‐(7H‐pyrrolo[2,3‐d]pyrimidin‐4‐ylamino)benzoate ( 2 ), were elucidated. The structure revealed two clear basic patches near the ATP pocket providing an explanation of TTBK1 for phosphorylation‐primed substrates. Interestingly, compound 2 displayed slow binding kinetics to TTBK1, the structure of TTBK1 in complex with this compound revealed a reorganization of the L199–D200 peptide backbone conformation together with altered hydrogen bonding with compound 2 . These conformational changes necessary for the binding of compound 2 are likely the basis of the slow kinetics. This first TTBK1 structure can assist the discovery of novel inhibitors for the treatment of Alzheimer’s disease.     

Mutation of recombinant catalytic subunit {alpha} of the protein kinase CK2 that affects catalytic efficiency and specificity

In order to understand better the structural and functionalrelations between protein kinase CK2 catalytic subunit, thetriphosphate moiety of ATP, the catalytic metal and the peptidicsubstrate, we built a structural model of Yarrowia lipolyticaprotein kinase CK2 catalytic subunit using the recently solvedthree-dimensional structure of the maize enzyme and the structureof cAMP-dependent protein kinase peptidic inhibitor (1CDK) astemplates. The overall structure of the catalytic subunit isclose to the structure solved by Niefind et al. It comprisestwo lobes, which move relative to each other. The peptide usedas substrate is tightly bound to the enzyme, at specific locations.Molecular dynamic calculations in combination with the studyof the structural model led us to identify amino acid residuesclose to the triphosphate moiety of ATP and a residue sufficientlyfar from the peptide that could be mutated so as to modify thespecificity of the enzyme. Site-directed mutagenesis was usedto replace by charged residues both glycine-48, a residue locatedwithin the glycine-rich loop, involved in binding of ATP phosphatemoiety, and glycine-177, a residue close to the active site.Kinetic properties of purified wild-type and mutated subunitswere studied with respect to ATP, MgCl₂ and protein kinase CK2specific peptide substrates. The catalytic efficiency of theG48D mutant increased by factors of 4 for ATP and 17.5 for theRRRADDSDDDDD peptide. The mutant G48K had a low activity withATP and no detectable activity with peptide substrates and wasalso inhibited by magnesium. An increased velocity of ADP releaseby G48D and the building of an electrostatic barrier betweenATP and the peptidic substrate in G48K could explain these results.The kinetic properties of the mutant G177K with ATP were notaffected, but the catalytic efficiency for the RRRADDSDDDDDsubstrate increased sixfold. Lysine 177 could interact withthe lysine-rich cluster involved in the specificity of proteinkinase CK2 towards acidic substrate, thereby increasing itsactivity.     

A Quantitative Comparison of Wild‐Type and Gatekeeper Mutant Cdk2 for Chemical Genetic Studies with ATP Analogues

Mutant kinase kinetics : Protein kinases with enlarged ATP binding sites are increasingly being used as tools to probe the functioning signal transduction cascades. Using human cyclin‐dependent kinase 2 as a model system, we demonstrate that enlargement of the ATP binding site does not substantially alter either the catalysis kinetics nor substrate or phosphorylation site selection.

     

Design and Analysis of the 4-Anilinoquin(az)oline Kinase Inhibition Profiles of GAK/SLK/STK10 Using Quantitative Structure-Activity Relationships

The 4-anilinoquinoline and 4-anilinoquinazoline ring systems have been the focus of significant efforts in prior kinase drug discovery programs, which have led to approved medicines. Broad kinome profiles of these compounds have now been assessed with the advent of advanced screening technologies. These ring systems, while originally designed for specific targets including epidermal growth factor receptor (EGFR), but actually display a number of potent collateral kinase targets, some of which have been associated with negative clinical outcomes. We have designed and synthesized a series of 4-anilinoquin(az)olines in order to better understand the structure-activity relationships of three main collateral kinase targets of quin(az)oline-based kinase inhibitors: cyclin G associated kinase (GAK), STE20-like serine/threonine-protein kinase (SLK) and serine/threonine-protein kinase 10 (STK10). This was achieved through a series of quantitative structure-activity relationship (QSAR) analysis, water mapping of the kinase ATP binding sites and extensive small-molecule X-ray structural analysis.     

Evaluation of the Binding Kinetics of RHEB with mTORC1 by In-Cell and In Vitro Assays

The mammalian/mechanistic target of rapamycin complex 1 (mTORC1) is activated by the small G-protein, Ras homolog enriched in brain (RHEB–GTPase). On lysosome, RHEB activates mTORC1 by binding the domains of N-heat, M-heat, and the focal adhesion targeting (FAT) domain, which allosterically regulates ATP binding in the active site for further phosphorylation. The crucial role of RHEB in regulating growth and survival through mTORC1 makes it a targetable site for anti-cancer therapeutics. However, the binding kinetics of RHEB to mTORC1 is still unknown at the molecular level. Therefore, we studied the kinetics by in vitro and in-cell protein–protein interaction (PPI) assays. To this end, we used the split-luciferase system (NanoBiT^®) for in-cell studies and prepared proteins for the in vitro measurements. Consequently, we demonstrated that RHEB binds to the whole mTOR both in the presence or absence of GTPγS, with five-fold weaker affinity in the presence of GTPγS. In addition, RHEB bound to the truncated mTOR fragments of N-heat domain (∆N, aa 60–167) or M-heat domain (∆M, aa 967–1023) with the same affinity in the absence of GTP. The reconstructed binding site of RHEB, ∆N-FAT-M, however, bound to RHEB with the same affinity as ∆N-M, indicating that the FAT domain (∆FAT, aa 1240–1360) is dispensable for RHEB binding. Furthermore, RHEB bound to the truncated kinase domain (∆ATP, aa 2148–2300) with higher affinity than to ∆N-FAT-M. In conclusion, RHEB engages two different binding sites of mTOR, ∆N-FAT-M and ∆ATP, with higher affinity for ∆ATP, which likely regulates the kinase activity of mTOR through multiple different biding modes.     

NMR backbone assignment of a protein kinase catalytic domain by a combination of several approaches: application to the catalytic subunit of cAMP-dependent protein kinase

Protein phosphorylation is one of the most important mechanisms used for intracellular regulation in eukaryotic cells. Currently, one of the best-characterized protein kinases is the catalytic subunit of cAMP-dependent protein kinase or protein kinase A (PKA). PKA has the typical bilobular structure of kinases, with the active site consisting of a cleft between the two structural lobes. For full kinase activity, the catalytic subunit has to be phosphorylated. The catalytic subunit of PKA has two main phosphorylation sites: Thr197 and Ser338. Binding of ATP or inhibitors to the ATP site induces large structural changes. Here we describe the partial backbone assignment of the PKA catalytic domain by NMR spectroscopy, which represents the first NMR assignment of any protein kinase catalytic domain. Backbone resonance assignment for the 42 kDa protein was accomplished by an approach employing 1) triply ((2)H,(13)C,(15)N) labeled protein and classical NMR assignment experiments, 2) back-calculation of chemical shifts from known X-ray structures, 3) use of paramagnetic adenosine derivatives as spin-labels, and 4) selective amino acid labeling. Interpretation of chemical-shift perturbations allowed mapping of the interaction surface with the protein kinase inhibitor H7. Furthermore, structural conformational changes were observed by comparison of backbone amide shifts obtained by 2D (1)H,(15)N TROSY of an inactive Thr197Ala mutant with the wild-type enzyme.     

Protein Contacts and Ligand Binding in the Inward‐Facing Model of Human P‐Glycoprotein

The primary aim of this work was to analyze the contacts between residues in the nucleotide binding domains (NBDs) and at the interface between the transmembrane domains (TMDs) and the NBDs in the inward‐open homology model of human P‐glycoprotein (P‐gp). The analysis revealed communication nets through hydrogen bonding in the NBD and at the NBD–TMD interface of each half involving residues from the adenosine triphosphate (ATP) motifs and the coupling helices of the intracellular loops. Similar networks have been identified in P‐gp conformations generated by molecular dynamics simulation. Differences have been recorded in the networking between both halves of P‐gp. Many of the residue contacts have also been observed in the X‐ray crystal structures of other ATP binding cassette (ABC) transporters, which confirms their validity. Next, possible binding pockets involving residues of importance for the TMD–NBD communication were identified. By studying these pockets, binding sites were suggested for rhodamine 123 (R‐site) and prazosin (regulatory site) at the NBD–TMD interface that agreed with the experimental data on their location. Additionally, one more R‐site in the protein cavity was proposed, in accordance with the available biochemical data. Together with the previously suggested Hoechst 33342 site (H‐site), all sites were interpreted with respect to their effects on the protein ATPase activity, in correspondence with the experimental observations. Several residues involved in key contacts in the P‐gp NBDs were proposed for further targeted mutagenesis experiments.     

Computational Insights into the Deleterious Impacts of Missense Variants on N-Acetyl-d-glucosamine Kinase Structure and Function

An enzyme of the mammalian amino-sugar metabolism pathway, N-acetylglucosamine kinase (NAGK), that synthesizes N-acetylglucosamine (GlcNAc)-6-phosphate, is reported to promote dynein functions during mitosis, axonal and dendritic growth, cell migration, and selective autophagy, which all are unrelated to its enzyme activity. As non-enzymatic structural functions can be altered by genetic variation, we made an effort in this study aimed at deciphering the pathological effect of nonsynonymous single-nucleotide polymorphisms (nsSNPs) in NAGK gene. An integrated computational approach, including molecular dynamics (MD) simulation and protein–protein docking simulation, was used to identify the damaging nsSNPs and their detailed structural and functional consequences. The analysis revealed the four most damaging variants (G11R, G32R, G120E, and A156D), which are highly conserved and functional, positioned in both small (G11R and G32R) and large (G120E and A156D) domains of NAGK. G11R is located in the ATP binding region, while variants present in the large domain (G120E and A156D) were found to induce substantial alterations in the structural organizations of both domains, including the ATP and substrate binding sites. Furthermore, all variants were found to reduce binding energy between NAGK and dynein subunit DYNLRB1, as revealed by protein–protein docking and MM-GBSA binding energy calculation supporting their deleteriousness on non-canonical function. We hope these findings will direct future studies to gain more insight into the role of these variants in the loss of NAGK function and their role in neurodevelopmental disorders.     

Ethanolamine kinase activity and compositions of diacylglycerols,phosphatidylcholines and phosphatidylethanolamines in livers of choline-deficient rats

These experiments were performed to find the reasons for the increased concentrations of docosahexaenoyl phosphatidylethanolamines (PE) in livers of choline-deficient rats. We measured the activity of ethanolamine kinase, which catalyzes the first step in PE formation. We also measured the compositions of PE and phosphatidylcholines (PC) and concentrations and fatty acid compositions of diacylglycerols (DG), which are precursors of PE. Young male rats were fed for one week a low-methionine, choline-deficient diet, or the same diet supplemented with choline. Ethanolamine kinase activity was measured in liver cytosol (100,000 g supernatant). Fatty acids were measured in total liver diacylglycerols and in microsomal PE and PC. Ethanolamine kinase activities were equal in choline-deficient and choline-supplemented rats. Concentrations of DG were elevated 6-fold by choline deficiency. The percentage of docosahexaenoic acid (22∶6n−3) in microsomal PE was nearly doubled by choline deficiency. Although the increased concentrations of PE in choline-deficient livers cannot be attributed to increased activity of ethanolamine kinase, the rate of PE formation probably was increased by increases in concentrations of its precursors, including DG. The disproportionate increase in 22∶6n−3 PE probably was caused by a selective formation of PE from DG that contain 22∶6n−3.     

Functional classification of protein kinase binding sites using Cavbase

Increasingly, drug-discovery processes focus on complete gene families. Tools for analyzing similarities and differences across protein families are important for the understanding of key functional features of proteins. Herein we present a method for classifying protein families on the basis of the properties of their active sites. We have developed Cavbase, a method for describing and comparing protein binding pockets, and show its application to the functional classification of the binding pockets of the protein family of protein kinases. A diverse set of kinase cavities is mutually compared and analyzed in terms of recurring functional recognition patterns in the active sites. We are able to propose a relevant classification based on the binding motifs in the active sites. The obtained classification provides a novel perspective on functional properties across protein space. The classification of the MAP and the c-Abl kinases is analyzed in detail, showing a clear separation of the respective kinase subfamilies. Remarkable cross-relations among protein kinases are detected, in contrast to sequence-based classifications, which are not able to detect these relations. Furthermore, our classification is able to highlight features important in the optimization of protein kinase inhibitors. Using small-molecule inhibition data we could rationalize cross-reactivities between unrelated kinases which become apparent in the structural comparison of their binding sites. This procedure helps in the identification of other possible kinase targets that behave similarly in "binding pocket space" to the kinase under consideration.