Bioinformatics Advance Access originally published online on May 23, 2006
Bioinformatics 2006 22(15):1846-1854; doi:10.1093/bioinformatics/btl199
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Imprint of evolutionary conservation and protein structure variation on the binding function of protein tyrosine kinases
Department of Pharmacology, University of California San Diego 9500 Gilman Drive, La Jolla, CA 92093-0392, USA
*To whom correspondence should be addressed.
 |
ABSTRACT |
|---|
 TOP ABSTRACT 1 INTRODUCTION2 SYSTEMS AND METHODS3 RESULTS AND DISCUSSION4 CONCLUSIONREFERENCES |
|---|
Motivation: According to the models of divergent molecular evolution, the evolvability of new protein function may depend on the induction of new phenotypic traits by a small number of mutations of the binding site residues. Evolutionary relationships between protein kinases are often employed to infer inhibitor binding profiles from sequence analysis. However, protein kinases binding profiles may display inhibitor selectivity within a given kinase subfamily, while exhibiting cross-activity between kinases that are phylogenetically remote from the prime target. The emerging insights into kinase function and evolution combined with a rapidly growing number of publically available crystal structures of protein kinases complexes have motivated structural bioinformatics analysis of sequence–structure relationships in determining the binding function of protein tyrosine kinases.
Results: In silico profiling of Imatinib mesylate and PD-173955 kinase inhibitors with protein tyrosine kinases is conducted on kinome scale by using evolutionary analysis and fingerprinting inhibitor–protein interactions with the panel of all publically available protein tyrosine kinases crystal structures. We have found that sequence plasticity of the binding site residues alone may not be sufficient to enable protein tyrosine kinases to readily evolve novel binding activities with inhibitors. While evolutionary signal derived solely from the tyrosine kinase sequence conservation can not be readily translated into the ligand binding phenotype, the proposed structural bioinformatics analysis can discriminate a functionally relevant kinase binding signal from a simple phylogenetic relationship. The results of this work reveal that protein conformational diversity is intimately linked with sequence plasticity of the binding site residues in achieving functional adaptability of protein kinases towards specific drug binding. This study offers a plausible molecular rationale to the experimental binding profiles of the studied kinase inhibitors and provides a theoretical basis for constructing functionally relevant kinase binding trees.
Contact: gverkhiv{at}ucsd.edu
Supplementary information: The supplementary material contains the details of the phylogenetic analysis of protein tyrosine kinases, including phylogenetic dendrograms of protein tyrosine kinases based on sequence alignments of the kinases catalytic domain and evolutionary conservation profiles of the binding site residues. This section provides a more detailed description of the Monte Carlo binding simulations, including energetic model and simulated tempering technique generalized for ligand–protein binding dynamics with the multiple protein tyrosine kinase structures.
|
1 INTRODUCTION |
|---|
|
TOPABSTRACT1 INTRODUCTION2 SYSTEMS AND METHODS3 RESULTS AND DISCUSSION4 CONCLUSIONREFERENCES |
|---|
Proteins with promiscuous functions may have divergently evolved to improvise novel or alter existing functions and to acquire higher specificity using sequence plasticity of only a small number of amino acids residing inside or near active sites (Aharoni et al., 2005; Copley et al., 2003; Gerlt et al., 2005; James and Tawfik, 2003; Yoshikuni, et al., 2006). The recent evidence from direct evolution experiments has indicated that substantial changes in the promiscuous functions of a protein need not come at the expense of its native function (Aharoni et al., 2005; James and Tawfik, 2003). Evolution of a new function may be driven by mutations that have little effect on the native function but large effects on the promiscuous functions. Functional promiscuity is often linked with protein conformational diversity, which are considered as evolvability traits that enable existing enzymes to rapidly evolve new activities (Aharoni et al., 2005; Copley et al., 2003; Gerlt et al., 2005; James and Tawfik, 2003; Yoshikuni, et al., 2006). The energy landscape view of molecular recognition (Kumar et al., 2000; Verkhivker et al., 2002)—whereby a protein sequence can adopt diverse structures in binding with different partners—provides a model framework whereby sequence conservation and structural diversity can be interlinked in deciphering how proteins can rapidly evolve new binding functions.
Comprehensive analysis of the protein kinase complement of the human genome (the ‘kinome’) (Hanks and Hunter, 1995; Hunter and Plowman, 1997; Kostich et al., 2002; Krupa and Srinivasan, 2002; Manning et al., 2002) has provided important insights into classification, evolution and function of human protein kinases, emerging as a major class of drug targets in recent years (Hubbard, 2002; Madhusudan and Ganesan, 2004). Evolutionary relationships between protein kinases (Manning et al., 2002) are often employed to infer inhibitor binding profiles from sequence analysis (Fabian et al., 2005). While protein kinases binding profiles may exhibit inhibitor selectivity within a given kinase subfamily, similar inhibitor activities are often displayed between phylogenetically distant targets (Brain et al., 2003). Imatinib mesylate (Gleevec, STI571 or CP57148B) (Supplementary Figure 1a) is highly effective in treating early-stage Chronic myelogenous leukemia (CML) acting as an effective direct inhibitor of ABL, KIT and PDGFR tyrosine kinases (Druker, 2004; Lydon and Druker, 2004; Deininger et al., 2005; Wong and Witte, 2004; Nagar et al., 2002). This inhibitor is a prime example of inherent complexity in inferring binding specificity by means of sequence homology of kinase domains, which assumes that phylogenetically similar kinases should display similar inhibitor binding profiles. The phylogenetic dendrogram based on sequence analysis of the protein kinases catalytic domain (Supplementary Figure 2) assigns PDGFR and KIT kinases as more divergent from the ABL tyrosine kinase than SRC subfamily. However, Imatinib mesylate exhibits comparable efficiencies and high selectivity at inhibiting PDGFR, KIT and ABL tyrosine kinases, while being largely ineffective in suppressing the tyrosine kinase activity of SRC family (Wong and Witte, 2004). The phylogenetic analysis of the protein tyrosine kinases would not have predicted selective inhibition of PDGFR, KIT and ABL tyrosine kinases by Imatinib mesylate without suggesting cross-binding to a broader range of closely related tyrosine kinases (Deininger et al., 2005; Wong and Witte, 2004). Hence, functional profile of Imatinib mesylate binding may not be directly linked with the position of protein kinases on the evolutionary dendrogram.
Pyrido [2,3-d]pyrimidine class of inhibitors, including PD-173955 (Supplementary Figure 1b), were initially developed as broadly active inhibitors of several tyrosine kinases, such as SRC, PDFGR and FGFR receptors (Hamby et al., 1997; Boschelli et al., 1998). In subsequent studies, these compounds were also shown to potently inhibit ABL and KIT tyrosine kinases (Wong and Witte, 2004; Nagar et al., 2002; Wisniewski et al., 2002). While sensitivity of the tyrosine kinases activity to the PD-173955 inhibitor may not be immediately evident from phylogenetic proximity, all these tyrosine kinases share the evolutionary conserved Thr residue at the gate-keeper site of the binding pocket. Interestingly, although the evolutionary profile of the Thr gate-keeper residue is conserved in all tyrosine kinases from the ABL and SRC families, the respective binding profile of Imatinib mesylate for ABL and SRC kinases can vary dramatically (Druker et al., 2004; Lydon and Druker, 2004: Deininger et al., 2005; Wong and Witte, 2004).
Although the catalytic domains of protein kinases in the active form are structurally very similar, crystallographic studies have revealed a remarkable plasticity and considerable kinase versatility in adopting vastly different and unique inactive conformations (Huse and Kuriyan, 2002), which arguably leads to high selectivity of Imatinib mesylate at inhibiting ABL, KIT and PDGFR tyrosine kinases. A significant body of experimental evidence suggests that protein kinases can occupy a range of natural conformations between inactive and active states and significant differences observed between structures of ABL kinase may be merely a function of the activation state of the ABL kinase (Schindler et al., 2000; Wisniewski et al., 2002; Nagar et al., 2002). In the framework of the energy landscape view of molecular recognition, both active and inactive protein conformations may exist in equilibrium which is shifted towards a preferable thermodynamic state upon ligand association.
In this work, we conduct in silico profiling of Imatinib mesylate and PD-173955 kinase inhibitors on kinome scale by using evolutionary analysis and fingerprinting inhibitor–protein interactions with the panel of all publically available protein tyrosine kinases crystal structures. Evolutionary history of the protein tyrosine kinases has led to the conservation patterns, which reflect not only general evolutionary substitution tendencies among amino acids, but are also influenced by the selective pressure on the binding site residues to fulfill structural integrity and functional specificity requirements. We set out to investigate how sequence plasticity of the binding site residues and protein kinase conformational diversity can be interrelated in achieving functional adaptability for specific drug binding.
|
2 SYSTEMS AND METHODS |
|---|
|
TOPABSTRACT1 INTRODUCTION2 SYSTEMS AND METHODS3 RESULTS AND DISCUSSION4 CONCLUSIONREFERENCES |
|---|
2.1 Structural classification of protein tyrosine kinases
All publically available crystal structures of protein tyrosine kinases in the Protein Data Bank (PDB) (Berman et al., 2000) are used to categorize the structural space into panels of active and inactive kinase conformations and to simulate Imatinib mesylate–kinase binding by interrogating inhibitor interactions with the respective ensembles of multiple protein structures. The 50 employed protein structures provide the following coverage of tyrosine kinase families : 6 ABL (1fpu, 1iep, 1m52, 1opj, 1opk, 1opl); 1 ACK1 (1u4d); 3 KIT (1pkg, 1t45, 1t46); 1 ZAP70 (1u59); 1 BTIK (1k2p); 1 ITK (1sm2); 1CSK (1byg); 2 SYK (1xbb, 1xbc); 2 EGFR (1m14, 1m17); 4 IGF1R (1m7n, 1k3a, 1jqh, 1p4o); 2 EphA2 (1mqb, 1jpa); 1 FAK (1mp8); 4 FGFR1 (2fgi, 1fgk, 1fgi, 1agw); 2 HGFR (1r0p, 1r1w); 1 FLT3 ( 1rjb); 3 HCK (1ad5, 1qcf; 2hck); 4 INSR (1irk, 1ir3, 1i44, 1p14); 5 LCK (1qpc, 1qpd, 1qpe, 1qpj, 3lck); 4 SRC (1fmk, 1ksw, 2ptk, 2src); 1 TIE2 (1fvr); 1 KDR/VEGFR (1vr2).
Among 50 protein tyrosine kinase crystal structures, the following 20 structures belong to the category of inactive kinase conformations : 3 ABL (1fpu, 1iep, 1opj); 2 KIT (1t45, 1t46); 1 BTIK (1k2p); 1CSK (1byg); 2 IGF1R (1m7n, 1k3a); 2 EphA2 (1mqb, 1jpa); 1 FAK (1mp8); 1 FGFR1 (1fgk); 1 HGFR (1r1w); 1FLT3 ( 1rjb); 1 HCK (1ad5); 1 INSR (1irk); 1 MUSK (1luf) ; 1 SRC (1fmk); 1 TIE2 (1fvr).
The superposition of all crystal structures within a protein tyrosine kinase family into a common reference frame is based on similarity of C
atoms for a common set of residues defining the ATP binding site in protein tyrosine kinases.
2.2 Monte Carlo binding simulations : energy model
The molecular recognition energetic model used in this study includes intramolecular energy terms, given by torsional and non-bonded contributions of the DREIDING force field (Mayo et al., 1990), and the intermolecular energy contributions calculated using the AMBER force field (Cornell et al., 1995) to describe protein–protein interactions combined with an implicit solvation model (Stouten et al., 1993). The dispersion–repulsion and electrostatic terms have been modified and include a soft core component that was originally developed in free energy simulations to remove the singularity in the potentials and improve numerical stability of the simulations (Beutler et al., 1994). Softening the AMBER force field can enhance sampling of the conformational space while retaining adequate description of the binding energy landscape.
A solvation term was added to the interaction potential to account for the free energy of interactions between the explicitly modelled atoms of the protein–protein system and the implicitly modelled solvent. The term was derived by considering the transfer of atom, from an environment where it is completely surrounded by solvent, to an environment in which it has explicit atomic neighbors. The solvent-accessible surface is created by rolling a spherical water probe with a radius of 1.4 A over the Van der Waals surface of the molecule. The center of the probe traces the solvent-accessible surface.
2.3 Monte Carlo binding simulations : simulated tempering dynamics
We have carried out equilibrium simulations with the ensembles of protein kinase conformations using parallel simulated tempering dynamics (Hansmann, 1997; Sugita and Okamoto, 1999; Verkhivker et al., 2001) with 50 replicas of the ligand–protein system attributed respectively to 50 different temperature levels that are uniformly distributed in the range between 5300 K and 300 K. In simulations with ensembles of multiple protein conformations, protein conformations are linearly assigned to each temperature level, that implies a consecutive assignment of protein conformations starting from the highest temperature level and allows each protein conformation from the ensemble at least once be assigned to a certain temperature level. Starting with the highest temperature, every pair of adjacent temperature configurations is tested for swapping until the final lowest value of temperature is reached. This process of swapping configurations is repeated 50 times after each simulation cycle for all replicas where the exchange of conformations presents an improved global update which increases thermalization of the system and allows regions with a small density of states to be sampled accurately. Monte Carlo moves are performed simultaneously and independently for each replica at the corresponding temperature level. After each simulation cycle, that is completed for all replicas, exchange of configurations for every pair of adjacent replicas at neighboring temperatures is introduced. The m th and n th replicas, described by a common Hamiltonian H(X1, ... , Xm, ... , Xn, ... , XN, are associated with the inverse temperatures ßm and ßn, and the corresponding conformations Xm and Xn. The exchange of conformations between adjacent replicas m and n is accepted or rejected according to Metropolis criterion with the probability
 | (1) |
 | (2) |
At equilibrium, the fraction of time that the ligand–protein system spends at a protein conformation 
= i to time spent at a protein conformation = j is determined by the Boltzmann distribution
 | (3) |
2.4 Binding free energy calculations
Binding free energies are computed using the molecular mechanics AMBER force field Cornell et al. (1995) and the solvation energy term based on continuum generalized Born and solvent accessible surface area (GB/SA) solvation model (Still et al., 1990; Mohamadi et al., 1990).
The binding free energy of the ligand–protein complex can be written as follows:
 | (4) |
 | (5) |
 | (6) |
In the GB/SA model, the Gcavity and Gvdw contributions are combined together via evaluating solvent-accessible surface areas:
 | (7) |
 | (8) |
|
3 RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACT1 INTRODUCTION2 SYSTEMS AND METHODS3 RESULTS AND DISCUSSION4 CONCLUSIONREFERENCES |
|---|
Evolutionary relationships in the protein tyrosine kinase family can be exploited by probing sequence conservation profiles of the binding site residues, the functional subset of the kinase residues, which are not necessarily near in sequence, but rather close in structural space (Vieth et al., 2005; Vulpetti and Bosotti, 2004). Phylogenetic classifications of the tyrosine kinase family and subalignments of the binding site residues were used to identify clusters of protein sequences with similar binding pockets. We have constructed the phylogenetic dendrogram based on the evolutionary conservation profile of the binding site residues defined from the crystal structures of the ATP complex with protein kinase A and Imatinib–mesylate complex with the ABL kinase (Supplementary Figure 3). Evolutionary relationships derived from the sequence alignment of the binding site residues are sensitive to the binding site definition and cannot discriminate a functionally relevant Imatinib mesylate binding phenotype unless binding site residues are defined from the crystal structure of the inhibitor with the ABL kinase (Supplementary Figure 3). In the latter case, phylogenetic tree can indeed display a functionally relevant close proximity of ABL, CKIT and PDGFR kinases, whereas ABL and SRC kinases appear less similar.
A simple mapping of the tyrosine kinases dendrogram with the structural space of known inactive and active tyrosine kinase conformations (Supplementary Figure 4) reveals a broad coverage of the tyrosine kinase phylogenetic footprint with the diverse family of protein kinases conformations which may have suggested a broader binding activity profile for Imatinib mesylate that has been actually observed. To provide a more robust functional mapping from sequence analysis to a structurally-driven binding phenotype, we conduct in silico profiling of kinase inhibitors by fingerprinting ligand–protein interactions with the panel of all publically available protein tyrosine kinases crystal structures. The results of simulations can be conveniently summarized by graphically mapping the phylogenetic footprint of tyrosine kinases with the respective densities of inactive and active protein tyrosine kinase structures which yield the low-energy complexes with the inhibitors (Figure 1).
In agreement with the experimental data, simulations with the ensemble of inactive tyrosine kinase conformations have predicted the unique binding mode of Imatinib mesylate from the crystal structure with the ABL kinase and a high degree of selectivity towards ABL and KIT kinases (Figure 2). The vast majority of Imatinib mesylate low-energy conformations reside within root mean square deviation (RMSD) = 2 Å from the crystallographic inhibitor conformation. In order to provide a more accurate characterization of Imatinib mesylate functional profile, the binding free energies are computed for all Imatinib mesylate conformations and the respective protein conformations obtained in simulations at T = 300 K. Interestingly, there is an appreciable correlation between small deviations within Imatinib mesylate crystallographic binding mode and energetic variations (Figure 3). A smaller peak of predicted bound conformations lying within RMSD = 0.5 Å from the crystal structure corresponds to the lowest binding free energies of –16 kcal/mol. The largest cluster of the predicted docked conformations deviates only by RMSD = 1.0 Å from the crystallographic inhibitor conformation. However, this subtle change is immediately reflected in the energetic peak shifted towards –15 kcal/mol and –14 kcal/mol (Figure 3). Overall, a high selectivity towards the inactive conformations of ABL and KIT is achieved via convergence to a narrow spectrum of bound conformations featuring the unique binding mode of Imatinib mesylate in the crystal structure of the ABL kinase complex.
Simulations with protein tyrosine kinases in the active form have revealed a significant structural departure of the low-energy inhibitor conformations from the Imatinib mesylate crystal structure featured in the ABL kinase complex (Figures 2 and 3). Furthermore, these bound conformations are distributed over a rather broad range of RMSD values, pointing to a multitude of alternative binding conformations for Imatinib mesylate with the active tyrosine conformations. The distribution of binding free energies is shifted towards energy values with a peak nearing –6 kcal/mol, which are considerably less favorable than the ones observed in simulations with the inactive tyrosine kinase structures (Figure 4). A subsequent structural analysis of the spectrum of Imatinib mesylate conformations yielding weak low-energy complexes with the active tyrosine kinase conformations have unveiled functionally relevant conformational variations around the alternative Imatinib mesylate binding mode (Supplementary Figure 5) originally observed in the crystal structure with the SYK kinase (Atwell et al., 2004).
The original concept that the inactive conformation of a protein kinase could present a structurally unique target has been debated recently, fuelled by revelations that Imatinib can also inhibit some other tyrosine kinases, In particular, the recent discovery of a novel binding mode for Imatinib mesylate in the complex with the SYK kinase provided a plausible structural model for explaining Gleevec inhibition of some activated tyrosine kinases (Atwell et al., 2004). Furthermore, a chemical proteomics approach applied to kinase profiling of Imatinib mesylate binding across a broad panel of kinases has identified that this inhibitor can also bind to the SRC-family tyrosine kinase LCK and to a few other members of SRC-family kinases, including FRK and FYN kinases (Fabian et al., 2005; Wissing et al., 2004). In agreement with these experimental data, the results of structural bioinformatics analysis have shown that only SYK and LCK active tyrosine kinases conformations are capable of accommodating fluctuations of Imatinib mesylate around its novel binding mode from the crystal structure with the SYK kinase (Figure 4, Supplementary Figure 5). Sequence conservation of the binding site residues interacting with Imatinib mesylate and conformational diversity of the inactive kinase states are interrelated in achieving functional adaptability towards specific Imatinib mesylate binding. The selectivity of Imatinib mesylate at inhibiting a small group of protein tyrosine kinases is achieved by the high specificity with which this inhibitor can recognize the unique structural motif of the activation loop present in the inactive conformation of the ABL and KIT kinases (Schindler et al., 2000; Wisniewski et al., 2002; Nagar et al., 2002).
In contrast, to Imatinib mesylate, PD-173955 binding profile is insensitive to the activation state of tyrosine kinases as both the phosphorylated and non-phosphorylated forms of ABL and SRC kinases yield low energy interaction energies with the inhibitor (Figure 2). The low-energy inhibitor conformations found in simulations with both active and inactive tyrosine kinases reside within RMSD = 2 Å from the crystal structure of PD-173955 in the complex with active ABL. In simulations with both active and inactive tyrosine kinase conformations, PD-173955 binding dynamics exhibits a consistent convergence to SRC, HCK, LCK, KIT and ABL tyrosine kinases and a high degree of structural similarity between the predicted unique PD-173955 binding mode and the crystal structure of the inhibitor in the ABL active kinase complex (Figure 2, Supplementary Figure 5). Binding free energy analysis has revealed that both active and inactive forms of ABL tyrosine kinases yield a rather similar distribution of favorable binding free energies (Figure 3). In agreement with the experimental data, the active form of the ABL kinase provides a slightly more favorable binding free energy for PD-173955 with a peak at –16 kcal/mol to –15 kcal/mol. In contrast, binding free energies for Imatinib mesylate with the active tyrosine kinase structures are significantly weaker (Figure 3). The experimental binding profiles have indicated that PD-173955 is more potent than Imatinib mesylate, inhibiting ABL kinase with an IC50 of about 5 nM and independent of the phosphorylation state, whereas Imatinib mesylate inhibits predominantly dephosphorylated form of the ABL kinase at 
100 nM (Druker, 2004; Lydon and Druker, 2004; Deininger et al., 2005; Wong and Witte, 2004). The results of structural bioinformatics analysis agree with the more favorable inhibition profile of PD-173955, effective against ABL kinase regardless of its phosphorylation state. Furthermore, these data provide structural and energetic rationale behind the original hypothesis that PD-173955 could inhibit ABL kinase more potently than Imatinib mesylate because it can bind with high binding affinity multiple forms of the kinase.
Structural stability of the inhibitor binding mode and evolutionary conservation of a small Thr residue at a critical site of the binding pocket appear to be sufficient to ensure PD-173955 adaptability towards multiple states of protein tyrosine kinase from SRC family. Interestingly, the recent chemical proteomics studies have identified several Ser/Thr kinases as highly sensitive pyrido[2,3-d]pyrimidine targets, although these protein kinases belong to different groups of the human kinome (Wissing et al., 2004). Nevertheless, the ‘space’ of Ser/Thr kinases and tyrosine kinases which bind PD-173955 inhibitor is ‘connected’ by a relatively small number of mutations in the binding site residues, all sharing the conserved Thr residue at the gate-keeper position which largely controls the mode of recognition and specificity profile.
We have found that sequence plasticity of the binding site residues alone may not be sufficient to enable protein tyrosine kinases to readily evolve binding activities with the diverse spectrum of kinase inhibitors. Protein conformational diversity is intimately connected with sequence plasticity of the binding site residues in achieving functional adaptability of tyrosine kinases towards high affinity inhibitor binding. While evolutionary signal derived solely from the tyrosine kinase sequence conservation can not be readily translated into the ligand binding phenotype, the proposed structural bioinformatics analysis can discriminate a functionally relevant kinase binding signal from a simple phylogenetic relationship.
Classical cross-activity analysis of kinase inhibitors is typically based on sequence analysis and assumption that evolutionary similar kinases on the kinase dendrogram should display similar inhibitor binding capabilities. However, kinase selectivity data may not be directly derived from structural or sequence relationships between kinase targets. The recently proposed chemogenomic classification of kinase space based entirely on small molecule selectivity data generally differs from sequence-based dendrogram analysis, but remains comparable for closely homologous targets (Vieth et al., 2005). However, while these trends may exist for some highly homologous kinases, the lack of high sequence identity between two kinases does not guarantee that inhibitor selectivity can be expected.
The present work provides support to the emerging evidence that sequence similarity among protein tyrosine kinases does not simply correlate with the inhibitor binding profile. Evolutionary conservation patterns in protein tyrosine kinases need to be considered in the context of structural similarity analysis of the inhibitor binding modes and classification of respective protein conformational changes which may lead to more robust metrics of functional annotation. With the rapid growth of structural and sequence databases, binding sites annotation of newly solved structures can be made either by homology to proteins with known binding sites or by classifying them within a ‘superfolds’ of analogous structures that have similar binding site locations. Functionally important annotations of binding specificity can be then attempted by combining multiple sequence alignments, evolutionary analysis of binding sites and in silico profiling of ligand–protein interactions on genomic scale.
|
4 CONCLUSION |
|---|
|
TOPABSTRACT1 INTRODUCTION2 SYSTEMS AND METHODS3 RESULTS AND DISCUSSION4 CONCLUSIONREFERENCES |
|---|
Evolutionary and structural aspects of kinase inhibitors binding are studied by interrogating ligand–protein interactions with the panel of all publically available protein tyrosine kinases structures, representing both the ensembles of active and inactive protein kinase conformations. The structural bioinformatics analysis reveals that sequence plasticity of the binding site residues is sampled together with protein conformational diversity in enabling protein kinases to evolve inhibitor binding activities. By detecting a functionally relevant kinase binding signal, the results of this work offer a molecular rationale to the experimental binding profiles of the inhibitors and provide a theoretical basis for constructing functionally relevant kinase binding trees.
|
|
|
|
|
FOOTNOTES |
|---|
Associate Editor: Anna Tramontano
Received on September 28, 2005; revised on April 14, 2006; accepted on May 18, 2006
|
REFERENCES |
|---|
|
TOPABSTRACT1 INTRODUCTION2 SYSTEMS AND METHODS3 RESULTS AND DISCUSSION4 CONCLUSIONREFERENCES |
|---|
Aharoni, A., et al. (2005) The ‘evolvability’ of promiscuous protein functions. Nature Genet, . 37, 73–76[Web of Science][Medline].
Atwell, S., et al. (2004) A novel mode of Gleevec binding is revealed by the structure of spleen tyrosine kinase. J. Biol. Chem, . 279, 55827–55832
Bain, J., et al. (2003) The specificities of protein kinase inhibitors: an update. Biochem. J, . 371, 199–204[CrossRef][Web of Science][Medline].
Berman, H.M., et al. (2000) The Protein Data Bank. Nucleic Acids Res, . 28, 235–242
Beutler, T.C., et al. (1994) Avoiding singularities and numerical instabilities in free energy calculations based on molecular simulations. Chem. Phys. Lett, . 222, 529–539[CrossRef][Web of Science].
Boschelli, D.H., et al. (1998) Synthesis and tyrosine kinase inhibitory activity of a series of 2-amino-8H-pyrido[2,3-d]pyrimidines: identification of potent, selective platelet-derived growth factor receptor tyrosine kinase inhibitors. J. Med. Chem, . 41, 4365–4377[CrossRef][Web of Science][Medline].
Copley, S.D. (2003) Enzymes with extra talents: moonlighting functions and catalytic promiscuity. Curr. Opin. Chem. Biol, . 7, 265–272[CrossRef][Web of Science][Medline].
Cornell, W.D., et al. (1995) A second generation force field for simulation of proteins, nucleic acids, and organic molecules. J. Am. Chem. Soc, . 117, 5179–5197[CrossRef][Web of Science].
Deininger, M., et al. (2005) The development of imatinib as a therapeutic agent for chronic myeloid leukemia. Blood, 105, 2640–2653
Druker, B.J. (2004) Imatinib as a Paradigm of Targeted Therapies. Adv. Cancer Res, . 91, 1–30[CrossRef][Web of Science][Medline].
Fabian, M.A., et al. (2005) A small molecule-kinase interaction map for clinical kinase inhibitors. Nat. Biotechnol, . 23, 329–336[CrossRef][Web of Science][Medline].
Gerlt, J.A., et al. (2005) Divergent evolution in the enolase superfamily: the interplay of mechanism and specificity. Arch. Biochem. Biophys, . 433, 59–70[CrossRef][Web of Science][Medline].
Hamby, J.M., et al. (1997) Structure-activity relationships for a novel series of pyrido(2,3-d]pyrimidine tyrosine kinase inhibitors. J. Med. Chem, . 40, 2296–2303[CrossRef][Web of Science][Medline].
Hanks, S.K. and Hunter, T. (1995) The eukaryotic protein kinase superfamily: kinase (catalytic) domain structure and classification. FASEB J, . 9, 576–596[Abstract].
Hansmann, U.H.E. (1997) Parallel tempering algorithm for conformational studies of biological molecules. Chem. Phys. Lett, . 281, 140–150[CrossRef][Web of Science].
Hubbard, S.R. (2002) Protein tyrosine kinases: autoregulation and small-molecule inhibition. Curr. Opin. Struct. Biol, . 12, 735–741[CrossRef][Web of Science][Medline].
Hunter, T. and Plowman, G.D. (1997) The protein kinases of budding yeast: six score and more. Trends Biochem. Sci, . 22, 18–22[Web of Science][Medline].
Huse, M. and Kuriyan, J. (2002) The conformational plasticity of protein kinases. Cell, 109, 226–282.
James, L.C. and Tawfik, D.S. Trends Biochem. Sci, . (2003) 28, 361–368[CrossRef][Web of Science][Medline].
Kostich, M., et al. (2002) Human members of the eukaryotic protein kinase family. Genome Biol, . 3, research0043.1-12.
Krupa, A. and Srinivasan, N. (2002) The repertoire of protein kinases encoded in the draft version of the human genome: atypical variations and uncommon domain combinations. Genome Biol, . 3, research0066.1-0066.
Kumar, S., et al. (2000) Folding and binding cascades: dynamic landscapes and population shifts. Protein Sci, . 9, 10–19[Web of Science][Medline].
Lydon, N.B. and Duker, B.J. (2004) Lessons learned from the development of imatinib. Leukemia Res, . 28, Suppl. 1, 29–38[CrossRef].
Madhusudan, S. and Ganesan, T.S. (2004) Tyrosine kinase inhibitors in cancer therapy. Clin. Biochem, . 37, 618–635[CrossRef][Web of Science][Medline].
Manning, G., et al. (2002) The protein kinase complement of the human genome. Science, 298, 1912–1934
Manning, G., et al. (2002) Evolution of protein kinase signaling from yeast to man. Trends Biochem. Sci, . 27, 514–520[CrossRef][Web of Science][Medline].
Mayo, S.L., et al. (1990) DREIDING: a generic force field for molecular simulation. J. Phys. Chem, . 94, 8897–8909[CrossRef][Web of Science].
Mohamadi, F., et al. (1990) MacroModel-an integrated software system for modeling organic and bioorganic molecules using molecular mechanics. J. Comput. Chem, . 11, 440–467[CrossRef][Web of Science].
Nagar, B., et al. (2002) Crystal structures of the kinase domain of c-Abl in complex with the small molecule inhibitors PD173955 and imatinib (STI-571). Cancer Res, . 62, 4236–4243
Schindler, T., et al. (2000) Structural mechanism for STI-571 inhibition of abelson tyrosine kinase. Science, 289, 1938–1942
Sugita, Y. and Okamoto, Y. (1999) Replica-exchange molecular dynamics method for protein folding. Chem. Phys. Lett, . 314, 141–151[CrossRef][Web of Science].
Still, W.C., et al. (1990) Semianalytical treatment of solvation for molecular mechanics and dynamics. J. Amer. Chem. Soc, . 112, 6127–6129[CrossRef][Web of Science].
Stouten, P.F.W., et al. (1993) An effective solvation term based on atomic occupancies for use in protein simulations. Mol. Simul, . 10, 97–120[CrossRef].
Verkhivker, G.M., et al. (2001) Parallel simulated tempering dynamics of ligand–protein binding with ensembles of protein conformations. Chem. Phys. Lett, . 337, 181–189[CrossRef][Web of Science].
Verkhivker, G.M., et al. (2002) Complexity and simplicity of ligand-macromolecule interactions: the energy landscape perspective. Curr. Opin. Struct. Biol, . 12, 197–203[CrossRef][Web of Science][Medline].
Vieth, M., et al. (2005) Kinomics: characterizing the therapeutically validated kinase space. Drug Discov. Today, 10, 839–846[CrossRef][Web of Science][Medline].
Vulpetti, A. and Bosotti, R. (2004) Sequence and structural analysis of kinase ATP pocket residues. Il Farmaco, 59, 759–765[CrossRef][Medline].
Wisniewski, D., et al. (2002) Characterization of potent inhibitors of the Bcr-Abl and the c-Kit receptor tyrosine kinases. Cancer Res, . 62, 4244–4255
Wissing, J., et al. (2004) Chemical proteomics analysis reveals alternative modes of action for pyrido[2,3-d]pyrimidine kinase inhibitors. Mol. Cell. Proteomics, 3, 1181–1193
Wong, S. and Witte, O.N. (2004) The BCR-ABL story: bench to bedside and back. Annu. Rev. Immunol, . 22, 247–306[CrossRef][Web of Science][Medline].
Yoshikuni, Y., et al. (2006) Designed divergent evolution of enzyme function. Nature, 439, .
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. M. Verkhivker Exploring sequence-structure relationships in the tyrosine kinome space: functional classification of the binding specificity mechanisms for cancer therapeutics Bioinformatics, August 1, 2007; 23(15): 1919 - 1926. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||