solutionconformationsanddynamicsofabl kinase ... · solutionconformationsanddynamicsofabl...

Solution Conformations and Dynamics of ABLKinase-Inhibitor Complexes Determined by NMRSubstantiate the Different Binding Modes ofImatinib/Nilotinib and Dasatinib*□S �

Received for publication, February 20, 2008, and in revised form, April 18, 2008 Published, JBC Papers in Press, April 22, 2008, DOI 10.1074/jbc.M801337200

Navratna Vajpai‡, Andre Strauss§, Gabriele Fendrich§, Sandra W. Cowan-Jacob§, Paul W. Manley§,Stephan Grzesiek‡1, and Wolfgang Jahnke§2

From the §Novartis Institutes for BioMedical Research, 4002 Basel, Switzerland and the ‡Biozentrum, University of Basel,4056 Basel, Switzerland

Current structural understanding of kinases is largely basedon x-ray crystallographic studies, whereas very little data existon the conformations and dynamics that kinases adopt in thesolution state. ABL kinase is an important drug target in thetreatment of chronic myelogenous leukemia. Here, we presentthe first characterization of ABL kinase in complex with threeclinical inhibitors (imatinib, nilotinib, anddasatinib) bymodernsolution NMR techniques. Structural and dynamical resultswere derived from complete backbone resonance assign-ments, experimental residual dipolar couplings, and 15Nrelaxation data. Residual dipolar coupling data on the ima-tinib and nilotinib complexes show that the activation loopadopts the inactive conformation, whereas the dasatinibcomplex preserves the active conformation, which does notsupport contrary predictions based upon molecular model-ing. Nanosecond as well as microsecond dynamics can bedetected for certain residues in the activation loop in theinactive and active conformation complexes.

Protein kinases play critical roles in intracellular signal trans-duction pathways, deregulation of which can lead to a variety ofpathological states and diseases such as cancer. These enzymesare therefore tightly regulated with multiple layers of control,including phosphorylation, myristoylation, and interactionwith SH23 and SH3 or other regulatory domains.Modulation ofkinase activity by therapeutic agents is a clinically validatedconcept, with many kinases considered to be attractive drug

targets. ABL kinase is such a target because the expression ofthe BCR-ABL fusion protein (caused by unfaithful repair ofDNA strand breaks in bone marrow hematopoietic stem cellsand subsequent t(9,22) chromosome translocation) leads tolife-threatening chronic myelogenous leukemia (1, 2). In BCR-ABL, the breakpoint cluster region BCR protein replaces theN-terminal autoregulatory domain of the Abelson ABL proteinto give a constitutively activated tyrosine kinase, which dereg-ulates signal transduction pathways, causing uncontrolled pro-liferation and impaired differentiation of progenitor cells.X-ray crystallography has revealed various active and inac-

tive conformational states of kinases, which are implicated intheir regulation and modulation by inhibitors (3). The activestates are characterized by certain conformations of the activa-tion loop, phosphate-binding loop (P-loop), and helix C, whichorient the catalytic machinery to phosphorylate substrates; inthe inactive states, one or more of these elements are in differ-ent conformations, such that substrate binding and/or catalysiscannot occur. An important determinant is the orientation ofthe conserved Asp-Phe-Gly motif within the activation loop.For efficient catalysis, this motif adopts a “DFG-in” conforma-tion. In contrast, the “DFG-out” conformation has this motifdisplaced from the orientation needed for binding the substrateATP to phosphorylate and activate downstream signaling pro-teins. Such a DFG-out conformation has been observed inmany inactive kinases, including ABL, IRK, KIT, and FLT3tyrosine kinases (4–7) as well as the serine/threonine kinasesp38 MAPK and BRAF (8, 9).Different kinase inhibitors can bind to and stabilize different

kinase conformations, as exemplified in Fig. 1 for different ABLinhibitors. Crystallographic studies have shown that the tyro-sine kinase inhibitor imatinib (Glivec�/Gleevec�), a highlyeffective treatment for chronic phase chronicmyelogenous leu-kemia (10), binds within the catalytic site of the inactive form ofABL with the activation loop in a DFG-out conformation (6,11–15). This conformation is very similar to that with nilotinib(16) (Tasigna�), a more potent and selective ABL inhibitordeveloped to inhibit imatinib-resistant mutant forms of BCR-ABL, which frequently emerge in advanced stages of chronicmyelogenous leukemia and lead to relapse and disease progres-sion (17). In contrast, crystallographic studies have shown thatthe multi-targeted ABL and SRC family kinase inhibitor dasat-

* This work was supported by Swiss National Science Foundation Grant31-109712 (to S. G.) and Novartis Pharma AG. The costs of publication ofthis article were defrayed in part by the payment of page charges. Thisarticle must therefore be hereby marked “advertisement” in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

� This article was selected as a Paper of the Week.□S The on-line version of this article (available at http://www.jbc.org) contains

supplemental Fig. 1.1 To whom correspondence may be addressed. E-mail: stephan.grzesiek@

unibas.ch.2 To whom correspondence may be addressed. E-mail: wolfgang.jahnke@

novartis.com.3 The abbreviations used are: SH, Src homology; MAPK, mitogen-activated

protein kinase; BisTris, 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymeth-yl)propane-1,3-diol; TCEP, tris(2-carboxyethyl)phosphine; RDC, residualdipolar coupling; NOE, nuclear Overhauser effect; HSQC, heteronuclearsingle quantum coherence.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 283, NO. 26, pp. 18292–18302, June 27, 2008© 2008 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

18292 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283 • NUMBER 26 • JUNE 27, 2008

by guest on February 11, 2019http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/cgi/content/full/M801337200/DC1

http://www.jbc.org/

inib (SPRYCEL�) (18) binds to the active DFG-in conformationof ABL (19). However, based onmolecular modeling (19, 20), ithas been hypothesized that dasatinib can also bind to the inac-tive DFG-out conformation, and even without experimentalsupport, this notion is becoming established. Crystallographicstudies have shown that other promiscuous kinase inhibitorsalso bind to active conformations of ABL, including com-pounds of the pyrido[2,3-d]pyrimidine class (21–23) such asPD180970 and the staurosporine derivative AFN941 (11). Aclear understanding of the physiologically relevant bindingmodes of BCR-ABL inhibitor complexes is of utmost impor-tance for the rational design of potent and selective inhibitorsthat can also counteract the emergence of drug-resistantmutant forms of BCR-ABL (12).Crystallographic analysis is limited to those biomolecular

states that crystallize, which may not capture the full ensembleof conformations that are available in solution under physiolog-ically more relevant conditions. These crystallographic statesmay be artificially stabilized by crystal contacts while highlydynamical parts of structures remain invisible. In principle,NMR spectroscopy can provide themissing characterization ofconformational ensembles and the dynamics of their intercon-

version. However, NMRwork on kinases has been severely lim-ited by their relatively large size, poor solubility, and the factthat they can often be produced only in expression systems thatdonot allow cost-effective labeling by 13C, 15N, and 2H isotopes.Only recently have a few NMR studies provided some limitedinsight on the dynamics of the Eph receptor (24) and p38MAPK (25) kinases fromchemical shift changes and line broad-ening effects.In this study, we have applied new techniques such as expres-

sion of isotopically labeled ABL kinase in baculovirus-infectedinsect cells and residual dipolar couplings, which provide pre-cise geometrical information, to characterize the solution con-formations and dynamics of theABL kinase domain in complexwith the three clinically used inhibitors: imatinib, nilotinib, anddasatinib.

EXPERIMENTAL PROCEDURES

Protein Expression and Purification—Expression and purifi-cation of uniform and amino acid-selective 13C/15N isotope-labeled ABL kinase in baculovirus-infected insect cells werecarried out as described previously using the construct His6-TEVsite-GAMDP-hABL(Ser229–Ser500) (26, 27).

FIGURE 1. Chemical structure of ABL kinase and its inhibitors. A, chemical structures of the compounds discussed in this work: imatinib, nilotinib, dasatinib,PD180970, and AFN941 (tetrahydrostaurosporine). B, ribbon diagram showing ABL kinase (gray) with nilotinib (green carbons) bound, highlighting the P-loop(red), the activation loop (magenta), and helix C (yellow). C, details of the conformations of the P-loop and activation loop with imatinib bound (yellow, ABL; gold,imatinib) (Protein Data Bank code 1IEP), with nilotinib bound (magenta, ABL; green, nilotinib) (Protein Data Bank code 3CS9), and with dasatinib bound (cyan,ABL; magenta, dasatinib) (Protein Data Bank code 2GQG). All three structures are shown in the same orientation based on superposition of the proteincoordinates. Residues highlighted in the activation loops are Asp381, Phe382, and Tyr393. In the P-loop, Glu255 is shown as an indication of the difference inconformation between the active and inactive ABL conformations.

Solution Conformations of ABL in Complex with Inhibitors

JUNE 27, 2008 • VOLUME 283 • NUMBER 26 JOURNAL OF BIOLOGICAL CHEMISTRY 18293


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Selectively [U-13C/15N]Phe-Gly-Met-Tyr (FGMY)- and[13CO]Leu-Thr-labeled ABL kinase was expressed in custom-made BioExpress 2000 medium containing [U-13C/15N]Phe-Gly-Met-Tyr and [13CO]Leu-Thr (Cambridge Isotope Labora-tories, Inc.) as described (28), but without addition of imatinibto the culturemedium. FGMY-labeledHis-ABL kinase was iso-lated by nickel affinity chromatography (nickel-nitrilotriaceticacid, Qiagen) with imidazole elution, yielding 20 mg of hetero-geneously phosphorylated kinase from two 0.5-liter cultures.Incubationwith AcTEVTM protease (100 units/mg of His-ABL;Invitrogen) and YOP protein-tyrosine phosphatase (1000units/ml of reaction; New England Biolabs) for 15 h at 6 °Cremoved the His tag and dephosphorylated the protein. Inhib-itor (imatinib, nilotinib, dasatinib, PD180970, or AFN941) wasadded to aliquots of the reaction from 20mM stock solutions inMe2SO. The ABL complexes were then purified by size exclu-sion chromatography (Superdex 75 HR10/30 column, GEHealthcare) in 20 mM BisTris, 150 mM NaCl, 1 mM EDTA, and3 mM TCEP (pH 6.5), except for the ABL-PD180970 complex,for which 20 mM Tris, 100 mM NaCl, 1 mM EDTA, and 2.5 mM

TCEP (pH7.6)was used. Purified complexeswere concentrated(Ultrafree-0.5, 10 kDa, Millipore) to 230–330 �M. Protein con-centration, purity, and stoichiometry were determined by highpressure liquid chromatography for each complex. Liquidchromatography/mass spectrometry analysis showed an incor-poration of 13C/15N label of 95% and 12% residual monophos-phorylation for the purified FGMY-labeled ABL kinase.NMR Samples—Uniformly 13C/15N- and 15N-labeled sam-

ples of ABL-imatinib complexes (1:1) were prepared as 0.4 mM

solutions in 250�l of 95%H2O and 5%D2O, 20mMBisTris, 100mM NaCl, 2 mM EDTA, and 3 mM dithiothreitol or TCEP (pH6.5). Selectively labeled samples of imatinib, nilotinib, dasat-inib, and PD180970 complexes (1:1) were prepared as solutions(0.32, 0.32, 0.22, and 0.22 mM respectively) in either 95% H2Oand 5% D2O, 20 mM BisTris, 150 mM NaCl, 2 mM EDTA, and 3mM dithiothreitol or TCEP (pH 6.5) (imatinib, nilotinib, anddasatinib) or 95%H2O and 5% D2O, 20 mM Tris, 100 mMNaCl,1 mM EDTA, and 2.5 mM TCEP (pH 7.6) (PD180970). Similarpreparations were tested for an ABL-AFN941 complex. How-ever, this complex precipitated even at the low concentration of0.1 mM, and no NMR data could be acquired. Non-isotropicsamples of selectively labeled imatinib and nilotinib (dasatinib)complexes were prepared by adding 30 mg/ml (20 mg/ml) fila-mentous phage Pf1 (Asla Biotech).NMR Resonance Assignments and Measurement of Residual

Dipolar Coupling (RDC) Values—NMR spectra were recordedat 293 K on Bruker DRX 600 MHz (with and without a Cryo-Probe) and 800 MHz (equipped with a TCI CryoProbe) spec-trometers. All spectrometers were equipped with triple-reso-nance, triple-axis pulsed-field gradient probes. Backboneassignments followed standard triple-resonance strategies withtwo- and three-dimensional experiments, including HNCO,HNCA, HN(CO)CA, and 15N-edited 1H-1H nuclear Over-hauser effect (NOE) spectroscopy. All NMR data were pro-cessed using the NMRPipe suite of programs (29) and analyzedwith NMRView (30) to obtain assignments. RDCs wereobtained as differences in the splitting observed in the two-

dimensional 1H-15N in-phase anti-phase experiments (31)under anisotropic and isotropic conditions.NMR Relaxation Experiments and Analysis—Standard 15N

relaxation measurements (T1/T2, {1H}-15N NOE) wererecorded on the ABL-imatinib complex (uniformly and selec-tively labeled samples) at 800 MHz. T1/T2 decay curves werefitted by an in-housewritten routine implemented inMATLAB(MathWorks, Inc.) using a simplex search minimization andMonte Carlo estimation of errors (see Fig. 7, A and B). Lipari-Szabo model-free analysis of 15N relaxation data was achievedusing the TENSOR2 suite of programs (supplemental Fig. 1)(32).

RESULTS

Resonance Assignment of the ABL-Imatinib Complex—As-signment of backboneNMRresonanceswas initially performedfor the ABL kinase domain (GAMDP-Ser229–Ser500, humanABL1, isoform 1A, 32 kDa) in its non-phosphorylated form andin complex with imatinib. We have shown previously that effi-cient production of well folded ABL with uniform 13C/15N iso-tope labeling is possible by the baculovirus Sf9 insect cellexpression system (26). NMR analysis of the ABL complex wasdifficult for two reasons. 1) Because of solubility problems, ABLconcentrations in the NMR samples had to be less than �0.4mM. 2) The assignment had to be carried out using protonatedprotein because cost-effective deuterium labeling is currentlynot possible in the insect cell system. Consequently, the shorttransverse relaxation times of the 32-kDa complex allowed onlyHNCO, HNCA, HNCOCA, and 15N-edited NOE spectroscopybackbone assignment experiments and prevented the use ofCBCA-type experiments, which would have yielded distinctiveamino acid-type information (33). Supplemental informationabout amino acid types was therefore obtained from a total of15 additional, selectively labeled ABL samples. Further detailson these samples and the obtained chemical shifts are describedelsewhere (34). The available assignments comprise 96% of allbackbone 1HN, 15N, 13C�, and 13COresonances, covering 254 ofthe 264 non-proline residues (Fig. 2). Unassigned residues con-sist of the N-terminal glycine, several residues within the acti-vation loop, and His361 lining the imatinib-binding surface.Line broadening of adjacent residues indicates that most of themissing residues are broadened beyond detection because ofintermediate conformational exchange.Design of Isotope Labeling Scheme for the Study of Various

Inhibitor Complexes—After fully assigning the resonances ofthe ABL-imatinib complex, a strategy employing selectiveamino acid labeling was devised for the rapid and unambiguousresonance assignment of key residues in the other inhibitorcomplexes. To select the best suited labeling scheme, residual1H-15Ndipolar couplingswere predicted based upon the crystalstructures of the various inhibitor complexes and the orienta-tion tensor of the ABL-imatinib complex measured in Pf1phages (35). Residues for selective labeling were then chosen tomaximize the differences of the predicted dipolar couplings inthe “inactive” DFG-out and “active” DFG-in conformations.Thus, maximal experimental differentiation by RDC databetween these two conformations should be achieved.




ww

.jbc.org/D

ownloaded from


http://www.jbc.org/

The chosen labeling scheme (FGMY) consists of uniform13C/15N labeling for Phe, Gly, Met, and Tyr and specific 13COlabeling for Thr and Leu residues. FGMY labeling covers fiveresidues (Gly249, Gly250, Gly251, Tyr253, and Gly254) in theP-loop and seven key residues (Phe382, Gly383, Ser385, Met388,Gly390, Tyr393, and Gly398) in the activation loop, with the Serlabeling being the consequence of metabolic scrambling of iso-tope-labeled glycine. The most important residues (Gly249,Met388, and Tyr393) are preceded by 13CO-labeled Leu or Thrsuch that they are distinctively detectable in an HNCO experi-ment. Besides the amino acid-type information, this selectivelabeling scheme had the advantage that all key resonances werecompletely free of overlap. The ultimate assignment of reso-nances was achieved by a combination of three- and two-di-

mensional versions of HNCA, HNCO, and HN(CO)CAexperiments.Chemical Shift Analysis—The chemical shift of a nucleus is a

sensitive probe of its local environment and can therefore serveas a fingerprint for different molecular conformations. Com-parison of the HSQC fingerprint spectra of the ABL-inhibitorcomplexes demonstrates that the two DFG-out complexes(imatinib and nilotinib) possess a high degree of similarity (Fig.3A), but are very distinct from the dasatinib DFG-in complex(Fig. 3B). The HSQC spectrum of the dasatinib complex has astrong resemblance to that of the PD180970 (DFG-in) complex(not shown), but the differences are larger than those betweenthe two DFG-out complexes as quantified by the average1H-15N chemical shift differences (��ave) shown in Fig. 3C.

6.57.58.59.510.511.5

105

W261A366 A407

K262E255E466 L301L248 D444

K294D421

I314Q300R239D363

H295Y257L302

W478L364R457

I418S410F486

L376L471A365

F425 L266V448T315

S417 V468K415Y435E409

K467 K291N479N297 T319

A288T306 E292G372 Y253

S348 Y440E308R367R362

N358G259 G463

T495 C369D455S446G249 W476

T277 Y413G442 G390

G383V304 S485

H375

S481G250G254

G426T240G398

G321

G303G251G436

K271

7.47.88.28.69.0

120

122

124

T267

L445

F283

A399Y342

R460V377

K245 F416 A474T243

V379L273

A487A225 F311

E462 D233 W430A269 A350D227 C464

V299I489E352 F382E453 A492

V339E258Q346

K247R307

E238V335

V280S500E286K285

A424Y353K356F317 L323

M278

L298

V256Y469E275

S420Y456 D241

L411

Y232W405M343

M472 R483E459

L452I313

K234

L324

E431V270

M437

D391K378I443M496V427E355

I242A344

E236Q252L429

Y326E499L340

A397

E281

A337 L341W235M388 L451Y393V268

S349L354

V371C305

L327

E373N331 N231

E282Q491

N414M226

M290

I432F493

N374 K357Y312

N368N322

A380

Y264

L428

E334

K454

K274

L370

125

115

118

A B

15N

1HN

ppm

G AM D P S P N Y D KW E ME R T D I T M K HK H K L G G GL G G G Q Y G E V Y E G VW K K Y S L T V A V K T L K E D T M E K E I K H P N L V QK E I K H P N L V Q230230 240240 250250 260260 270270 280280 290290

L L G V C T R E P P F Y I I T E F M T Y G N L L D Y L R E C N R Q E V N A V V L L Y M A T Q I S S A M E Y L E K K N F I H R D L A A R N C L V G E N H T Q I S S A M E Y L E K K N F I H R D L A A R N C L V G E N H310310 320320 330330 340340 350350

L V K V A D F G L S R L M T G D T YT Y T A H A GA G A K F P380380 390390 400400

I K W T A P E S L A Y N K F SW T A P E S L A Y N K F S410410

300300

360360 370370

I K S D V W A F G V L L W E I A T Y G M S P Y P G I D L S Q V Y

L L E K D Y R M E R P E G C P

420420 430430 440440

460460E

450450E K V Y E L M R A C W Q W N P S D R P S F A E I H Q A F E T M F Q E S

470470 480480 490490 500500

Activation loop

P-loop

V260

L284

V E E F L K A A V M V E E F L K A A V M

C-helixC-helix

* * **

FIGURE 2. A, 1HN-15N HSQC-transverse relaxation optimized spectrum of uniformly 15N-labeled ABL-imatinib complex with assignment of resonances.Asterisks indicate unassigned Trp� side chain resonances. B, enlarged region of the box shown in A. The amino acid sequence of the ABL kinase domain(GAMDP-Ser229–Ser500) with its secondary structure and the mentioned activation loop (magenta), P-loop (red), and helix C (gold) is shown at the top.Boldface helices indicate �-helices, whereas dotted helices indicate 310-helices. Unassigned residues are underlined. The spectrum was acquired for 1 hat a sample concentration of �0.4 mM.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

∆δave [ppm]

Activation loopActivation loopP-loopP-loop HingeHinge

7.0

G254

S481

G250

S485G383

G390

S446

S348

M226

S229

F493

F359S438

Y253

S385S265

Y435

M290

Y413

M351

Y440

M458

S349

M437M496

M472

Y469

Y232Y456

M388

Y326

S500Y353

F382

S410M278S420

F311

F283Y342

F416

107

111

115

119

123

127

S481

S485

G251

G383 / G398 G250

S446G390 M351

Y413

Y440

M458M290Y435

S265

S348

M226

S229F493

F359

Y253

S438

Y449F497

M388

F311

Y342

F382

F283

Y469

Y232

M278

Y456

S500S410

S349Y326

M496

M237

M472

S420 Y353

7.07.58.08.57.58.08.5 ppm1HN

15N**

*

*

*

*

**

*

*

*

**

*

F497

Y449

A B

Y393

Y320

Y320

C

3

Y3

*6

9

0

0.5

1Imatinib - NilotinibImatinib - DasatinibDasatinib - PD180970

220 240 260 280 300 320 340 360 380 400 420 440 460 480 500Residue number

250 251

253283

290

317321

383

388

390398

FIGURE 3. Chemical shift analysis of selectively labeled (FGMY) ABL-inhibitor complexes. A and B, extracted region from the 1HN-15N HSQC spectra ofselectively FGMY-labeled ABL kinase in complex with imatinib (A; black) and dasatinib (B; black). Resonances of the ABL-nilotinib complex (green) are shown forcomparison. Residues labeled in red show the largest chemical shift changes. Asterisks indicate unassigned resonances of a low molecular mass impurity.C, weighted chemical shift differences ��ave � (��2(N)/50 � ��2(H)/2)1⁄2 between imatinib and nilotinib (inactive-inactive; black), imatinib and dasatinib(inactive-active; red), and dasatinib and PD180970 (active-active; blue) complexes. Open circles for Gly383/Gly398 indicate ambiguous assignments for dasatinib.Boxes at the top indicate secondary structure elements showing �-sheets (black) and helices (white). The spectra were acquired for 6 h at a sample concentra-tion of �0.3 mM (�0.2 mM) for imatinib and nilotinib (dasatinib) complexes.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Between imatinib and nilotinib, chemical shift differenceslarger than 0.1 ppm are detected only for Met290, Phe317, andGly383, which are in direct contact with the inhibitors. Muchstronger differences are observed between the dasatinib andimatinib complexes in the region of the P-loop around Gly250,

the hinge region residues Phe317 and Gly321, and the activationloop around Met388. Whereas the residues in the P-loop andhinge region participate in direct interactions with the inhibi-tor, the affected residues in the activation loop (Met388 andGly390) are not in direct contact, and hence, their chemical shiftchanges are in agreement with an allosteric reorientation of theactivation loop (Fig. 4).Residual Dipolar Couplings and Solution Structures—A

more quantitative description of the conformations in the dif-ferent inhibitor complexeswas obtained byRDCs. These can beinduced in solution by the weak alignment of biomacromol-ecules (36) and provide ameasure of the orientation of internu-clear vectors with respect to a fixed coordinate system. Thus,the RDC of the amide N-H bond (1DNH) is given as follows:1DNH � 1DNH,max(P2(cos�) � �/2sin2�cos2�), where 1DNH,maxis a constant depending on the degree of orientation, P2 is thesecond Legendre polynomial, � is the rhombicity of the align-ment tensor, and � and � are polar coordinates of the N-Hvector in the principal axis system of the alignment tensor (36).Because RDCs can be measured with high precision, their geo-metric dependence makes them a powerful tool to study solu-tion conformations and compare them with other structuralmodels such as solid-state x-ray crystal structures.Weak alignment of the selectively labeled ABL-inhibitor

complexes was achieved by the addition of filamentous bacte-riophage Pf1 (35). Large 1DNH RDCs (�30 Hz) were obtainedfor the imatinib, nilotinib, and dasatinib complexes, which indi-cated substantial alignment and allowed for high sensitivitydetection. For the PD180970 complex, the spectral quality wasinsufficient, and several key resonances were unobservablebecause of intermediate conformational exchange.For the imatinib and nilotinib complexes (DFG-out), the

RDC values are strikingly similar throughout the protein (Fig.5A), implying very similar solution structures and dynamics for

FIGURE 4. Mapping of largely shifted residues on three-dimensionalstructures of ABL-inhibitor complexes. Overlaid structures of ABL-inhibitorcomplexes are shown. A, ABL-imatinib (gray; Protein Data Bank code 1IEP)and ABL-nilotinib (yellow; code 3CS9). The P-loop, activation loop, and inhib-itors are colored in blue and green for imatinib and nilotinib complexes,respectively. B, ABL-imatinib (gray; code 1IEP) and ABL-dasatinib (green; code2GQG, molecule B). The P-loop, activation loop, and inhibitors are colored inblue and yellow for imatinib and dasatinib complexes, respectively. Residuesindicated in red show relatively larger changes in the chemical shifts (�0.1ppm for A and �0.3 for B).

-30

-20

-10

0

10

20

30

RD

C H

N-N

[Hz]

220 240 260 280 300 320 340 360 380 400 420 440 460 480 500Residue number

-30

-20

-10

0

10

20

30

40


A

B

ImatinibNilotinib

ImatinibDasatinib

FIGURE 5. Analysis of RDCs (measured in Pf1 phages) of selectivelylabeled (FGMY) ABL-inhibitor complexes. Shown is a comparison of theexperimental 1HN-15N RDCs obtained for imatinib and nilotinib complexes (A)and imatinib and dasatinib complexes (B) shown along the primary sequence.Open circles for Gly383/Gly398 indicate ambiguous assignments for dasatinib.Error bars indicate variances of two independently measured data sets. Boxesat the top indicate secondary structure elements: �-sheets (black) and helices(white).




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

both complexes. In marked contrast, RDCs for the dasatinibcomplex (Fig. 5B) differ substantially from those for the ima-tinib and nilotinib complexes in both the activation loop andthe P-loop. Thus, the solution conformation of the dasatinibcomplex clearly differs from that of the imatinib and nilotinibcomplexes, which corroborates the results of the chemical shiftanalysis.To characterize the conformations of the activation loop

(Asp381–Pro402), P-loop (Lys247–Glu255), and hinge region(Phe317–Leu323) in detail, theoretical RDC values were calcu-lated for each complex. For this, alignment tensors were deter-mined employing a linear fit procedure (37) using the respec-tive crystal structures and the measured RDCs, but excludingthe activation loop, P-loop, hinge region, and the flexible resi-dues at the N terminus (�Met237) and C terminus (�Phe493).Using these alignment tensors together with the crystal coordi-nates, RDC values were predicted for the entire protein, withthe previously excluded regions included. These theoreticalvalues were then compared with the experimental RDC valuesfor the imatinib, nilotinib, and dasatinib complexes (Fig. 6).For the imatinib and nilotinib complexes (Fig. 6, A and B), it

is evident that, besides the flexible N-terminal region, all RDCvalues throughout the entire protein including the loop regionsare in perfect agreement with the crystal structures. In partic-ular, this is the case for Phe382, Gly383, Ser385, Met388, Gly390,Tyr393, and Gly398 in the activation loop. This demonstratesthat the ABL-imatinib and ABL-nilotinib complexes adopt the

inactive DFG-out conformation insolution and that any dynamic vari-ations from the crystal structurecoordinates must be small.Taking into account the slightly

larger experimental errors for thedasatinib complex, the agreementbetween measured RDC values andthose predicted from the crystalstructure (Protein Data Bank code2GQG) is also very good for thiscomplex. The asymmetric unit ofthe crystal structure 2GQG con-tains two ABLmolecules, one in thephosphorylated form (molecule A;phospho-Tyr393) and one in thenon-phosphorylated form (mole-cule B). Both structures have almostidentical backbone conformations.Predictions are shown in Fig. 6Cfor the non-phosphorylated formbecause the ABL protein used forsolution NMR was also non-phos-phorylated. Besides the flexible Nand C termini, moderate deviationsbetween measured and predictedRDCs outside of the experimentalerror are observed only for turn res-idues Gly250, Phe311, Met437, Gly442,andGly463, which are all part of loopregions. For all other unambiguous

assignments, very close agreement is observed. In particular,Phe382, Met388, and Gly390 within the activation loop could bedetected and assigned unambiguously. Because of exchangebroadening (see below), a further glycine resonance could onlybe assigned in an ambiguous way either to Gly383 or Gly398(shown as open circles in Fig. 6, C–F). However, for both theunambiguous and the two possible ambiguous assignments inthe activation loop, the experimental RDCs correspond veryclosely to the prediction of the 2GQG structure. Very similaragreement is found when the experimental data are comparedwith the crystal conformation of the dasatinib complex withphosphorylated ABL (2GQG, molecule A) (data not shown).Thus, we conclude that, in solution, the activation loop of thedasatinib complex predominates in the active DFG-in confor-mation corresponding to that of the 2GQG crystal structure.To estimate the discriminative power of RDC values for dif-

ferent conformations of the activation loop, we compared theexperimental RDCs of the dasatinib complex with those pre-dicted from both the “inactive state” protein in complex withimatinib (Fig. 6D) and the “active state” protein in complexwiththe structurally unrelated ABL kinase inhibitors PD180970(Fig. 6E) and AFN941 (Fig. 6F). For the inactive state imatinibcomplex, all of the predictions for the activation loop region areoutside of the error limits of the measured RDCs, showing thatthe dasatinib complex does not sample the inactive imatinibconformation to a significant extent.

-30-15

01530

-30-15

01530

-30-15

01530

-30-15

01530

RD

C H

N [H

z]

-30-15

01530

220 240 260 280 300 320 340 360 380 400 420 440 460 480 500Residue number

-30-15

01530


398

388

A

F

E

D

B

C

442437

311250

463

Imatinib 1IEP

Nilotinib 3CS9

Dasatinib 2GQG:B

Dasatinib 1IEP

Dasatinib 2HZI

Dasatinib 2HZ4

382390

383

FIGURE 6. Predicted versus experimental RDCs. The RDC values predicted from the crystal structures areshown in red, and the experimental RDC values are shown in black. A, ABL-imatinib (Protein Data Bank code1IEP); B, ABL-nilotinib (code 3CS9); C, ABL-dasatinib (code 2GQG, molecule B); D, ABL-dasatinib (code 1IEP);E, ABL-dasatinib (PD180970; code 2HZI); F, ABL-dasatinib (AFN941; code 2HZ4). Open circles for Gly383/Gly398

indicate ambiguous assignments for dasatinib. The alignment tensor was calculated using the experimentalRDCs of the complexes mentioned above and the coordinates taken from the crystal structures indicated inparentheses.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

In the fast chemical exchange regime, the RDC values of anaveraging ensemble of structures are given by the average overthe RDCs of the different conformations. Using an error esti-mate of 3 Hz for both experimental and predicted data, wecalculate from the large differences and, in some cases, the dif-ferent sign between measured and predicted data (e.g. forPhe382 and Gly383) that we would detect an inactive imatinibconformation if it were populated by more than �15%. How-ever, our experimental data are fully compatible with the ABL-dasatinib complex being exclusively in the active conformation.The crystal structures of the PD180970 and AFN941 com-

plexes show that these inhibitors also bind to ABL in the activeconformation, although the path of the activation loop in thesecomplexes is more variable than in the dasatinib complex.Thus, the PD180970 complex is in the active conformationwithrespect to most of the activation loop, but the crystal structurefor this complex shows the DFG motif in a conformation inwhich the Asp side chain is flipped over to form a hydrogenbond with a main chain carbonyl group. This conformationdoes not support the binding of ATP, yet is completely differentfrom the DFG-out conformation (11). The experimental RDCvalues for the dasatinib complex strongly deviate in the DFGregion (positions 381–383) from the predicted RDCvalues (Fig.6E). Therefore, we conclude that the PD180970 crystal DFGconformation is not highly populated in solution by the dasat-inib complex. In contrast, the predictions from the ABL-AFN941 complex (Fig. 6D), which has a typical active DFG-inconformation similar to 2GQG, are closer to those from thedasatinib structure and the experimental RDCs in the activa-tion loop. However, a strong deviation is observed for Gly254 inthe P-loop region, probably because the crystal structure shows

disorder in this region and only partof the P-loop could be seen in theelectron density.In summary, all RDC data indi-

cate that for the imatinib and nilo-tinib complexes, the ensembles ofsolution conformations are veryclose to the static structuresobserved in the crystal. The RDCdata also unambiguously show thatthe conformational ensemble of thedasatinib complex in solution clus-ters around the active DFG-in con-formation observed in the crystaland that inactive DFG-out confor-mations are not sampled to a signif-icant extent.Backbone Dynamics—The back-

bone of the ABL-imatinib complexwas characterized by 15N relaxationexperiments (Fig. 7). Decreases in15N T1 and {1H}-15N NOE valuesand an increase inT2 values at the Nterminus beforeMet237 and at the Cterminus after Phe493 indicate highnanosecond mobility at both ter-mini. A high rigidity throughout

most of the remaining residues is evident from rather uniform15N T1 and T2 values and {1H}-15N NOE values close to 0.8. Aclear exception is the activation loop, which has {1H}-15N NOEvalues close to 0.5, together with decreasedT1 and increasedT2values, indicating large amplitude motions on the subnanosec-ond time scale. Further regions of higher mobility can be iden-tified around Glu462 and close to Lys274, which are both locatedin turns and have very high temperature factors in the crystalstructures. Notably, the P-loop (Lys247–Glu255) of the ABL-imatinib complex does not show pronounced variations in 15NT1, T2, or {1H}-15N NOE values, which would indicate highsubnanosecond mobility. For the dasatinib complex, disorderwas detected in this region by x-ray crystallography (19).Because of the low concentrations and the selective labeling,quantitative 15N relaxation data on the latter complex are cur-rentlymissing.However, stronger conformational exchange forthe dasatinib complex than for the imatinib complex may beinferred from the detected line broadening in this region (seebelow).A quantitative evaluation of the relaxation data was carried

out by the program TENSOR2 (32), yielding the isotropic rota-tional correlation time (c), the Lipari-Szabo (38), subnanosec-ond order parameters (S2), and exchange contributions totransverse relaxation from conformational exchange on themicro- to millisecond range (supplemental Fig. 1). The result-ing value for c of 21 ns is slightly larger than expected for amolecule of the size of ABL tumbling in aqueous solution at20 °C. Presumably, this larger value is caused by the onset ofaggregation, which occurs at the concentration of 0.4 mM usedfor the experiments. However, further dilution was notattempted because of the reduced sensitivity that would result.

1.5

3.0

4.5

T1

[s]

0

40

80

120

160

T2

[ms]

220 240 260 280 300 320 340 360 380 400 420 440 460 480 500Residue number

-0.4

0

0.4

0.8

{1 H}-

15N

NO

EActivation loopActivation loopP-loopP-loop HingeHinge

A

B

C

FIGURE 7. 15N relaxation data of the ABL-imatinib complex. T1 (A), T2 (B), and hetero-{1H}-15N NOE (C) valuesare shown along the primary sequence. Boxes at the top indicate secondary structure elements: �-sheets(black) and helices (white). For clarity, error bars were omitted. Typical errors propagated from the experimentalnoise by Monte Carlo estimates are as follows: T1 � �250 ms; T2 � �3 ms; and NOE � �0.03.




ww

.jbc.org/D

ownloaded from


http://www.jbc.org/

The higher mobility in the region of the activation loop isreflected in the TENSOR2 analysis both by a reduced orderparameter (S2) of �0.6–0.7 and contributions from chemicalexchange broadening on the microsecond time scale of up to25–30 Hz.Such chemical exchange effects are clearly visible as a broad-

ening and weakening of the resonance lines in the 1H-15NHSQC spectra. Fig. 8 shows the respective peaks of the activa-tion loop and P-loop for the imatinib, nilotinib, and dasatinibcomplexes. For imatinib, prominent line broadening isobserved for Met388 and Tyr393, which cannot be attributed tohydrogen exchange with water because exchange peaks withwater are absent in the 15N-editedNOE spectra. Hence, the linebroadening is caused by conformational exchange on themicrosecond time scale of chemical shifts. The broadening ofTyr393 is particularly interesting because this residue becomesphosphorylated in the activated complex. Similar line broaden-ing is observed for the nilotinib complex. However, muchmorepronounced broadening occurs for the dasatinib complex, e.g.Ser385, and Tyr393 could not be detected at all. This clearly indi-cates a differing dynamic behavior of the activation loop in thedasatinib complex.Within the P-loop, weak exchange broadening is observed

for Gly249 and Tyr253 in the case of imatinib and nilotinib (Fig.8). Again, much stronger exchange broadening occurs forGly249 in the dasatinib complex. Thus, also the P-loop appearsmore mobile in the case of the active state inhibitor dasatinib.These relaxation results on the ABL-inhibitor complexes are

significant because they directly show the presence of dynamicprocesses in several regions of the protein, including the acti-vation loop. It should be noted, however, that although theeffects of line broadening are considerable, they do not neces-sarily imply that the populations of the other conformations,which are exchanging with main species, are large. Admixturesof populations on the order of 1% can lead to significant broad-ening effects (39). Thus, the detection of such motions by linebroadening is not contradicting the finding from the RDC anal-ysis that the major part of the solution ensemble is close to thecrystal structure.

DISCUSSION

Our results compose the first detailed structural character-ization of protein kinase-inhibitor complexes in solution. They

are based on the availability of thealmost complete assignment of thebackbone resonances of the ABL-imatinib complex and partialresonance assignments of the ABL-nilotinib, ABL-dasatinib, andABL-PD180970 complexes. Thisallowed an analysis of chemical shiftperturbations, RDC, and 15N relax-ation data. For the imatinib andnilotinib complexes, the ensembleof solution conformations closelyresembles the static inactive DFG-out structure determined in thecrystal, although residual mobility

of the activation loop can be detected from 15N relaxation dataand the line broadening of some resonances.For the dasatinib complex, the RDC data clearly show that

the ensemble of solution conformations is close to the activeDFG-in structure. However, line broadening effects aroundSer385 andTyr393 in the activation loop andGly249 in the P-loopindicate the presence ofmicrosecond tomillisecondmotions inthese regions. Relaxation dispersion can reveal the time scaleand chemical shift differences of the species involved in theexchange (39). Because of the low solubility (0.2 mM) of thedasatinib complex and the weak intensity of the signals, suchexperiments were not attempted. Nevertheless, based on theclose agreement between measured RDCs and the predictionaccording to the active state conformation, the amplitude ofthese microsecond to millisecond motions and/or the popula-tions of the exchanging minor conformations should be small.Such minor conformations may be the result of small rear-rangements of the backbone or the side chains or variations inhydrogen bond geometries.Based onmolecular modeling andmolecular dynamics stud-

ies, it has been hypothesized that dasatinib can bind to both theactiveDFG-in and inactiveDFG-out conformations ofABL (19,20). This notion has become widespread despite the absence ofsupportive experimental evidence and the fact that the x-raystructure of the ABL-dasatinib complex shows only the activeABL conformation. This study is the first to actually assess theextent ofDFG-out conformation in theABL-dasatinib complexin solution. No significant admixture of the DFG-out confor-mation is detectable from the measured RDC values. In a fur-ther experiment (data not shown), we displaced imatinib byadding dasatinib in high excess to the ABL-imatinib complexrather than adding dasatinib directly to unliganded ABL. Evenwhen offering the preformed inactive DFG-out state to dasat-inib in this manner, the resulting ensemble of conformations isindistinguishable from the ensemble observed when addingdasatinib to unliganded ABL.This study was performedwith non-phosphorylated protein,

which was necessary to allow the protein to also adopt the inac-tive DFG-out conformation. The phosphorylation of Tyr393 inthe activation loop stabilizes the active conformation of theprotein by forming interactions with neighboring side chains(40). It can be expected that this will reduce the flexibility of thedasatinib complex, thereby narrowing the ensemble of the

F382 G383 S385 M388 G390 Y393 G398+ - - - - ++ - -

G249 G250 G251 Y253 G254

Imatinib

Nilotinib

Dasatinib

A

B

C

+ - + + - + - + -ImatinibH2O exch.

P-loop Activation loop

FIGURE 8. Evidence of chemical exchange broadening in the P-loop and activation loop of ABL kinasecomplexes. The panels show small regions of the 1HN-15N HSQC spectra of selectively FGMY-labeled ABLextracted at sequential residue positions of the P-loop and activation loop for the imatinib (A), nilotinib (B), anddasatinib (C) complexes. Empty boxes indicate absence of the resonance in the respective spectrum. The infor-mation at the top indicates the strength of amide proton exchange (exch.) with water as detected by theintensity of cross-peaks to water in the 15N-edited NOE spectrum.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

active DFG-in conformations even further. For the same rea-son, any propensity of the ABL-dasatinib complex to adopt theinactive DFG-out conformation should be even more reduced.It is generally believed that there is only one active state,

which satisfies the requirement to have all essential elementscorrectly orientated for efficient catalysis. In contrast, multipleinactive states may exist. This is supported by experimentalfindings for both ABL and SRC kinases in complex with variousligands (41). Our findings of very similar inactive states for theimatinib and nilotinib complexes do not contradict this notion;although multiple ligand-dependent inactive states may exist,we have shown that the inactive state with a particular ligand iswell defined and resembles closely the conformation observedin the crystal. However, these inactive conformations are notcompletely rigid because we still observed high nanosecondbackbone flexibility within the activation loop despite the factthat the RDC values indicate that the ensemble average is closeto the x-ray structure.The observed flexibility seen in both the active and inactive

states is likely to be an intrinsic requirement for catalytic activ-ity and for the transition between the active and inactive con-formations of ABL as well as other kinases (42). Indeed, molec-ular dynamics calculations have shown that the various inactiveABL kinase conformations may be necessary intermediates inthis transition (41).Considering the tendency of different kinases to adopt differ-

ent inactive states, e.g. either DFG-in or DFG-out conforma-tions, the free energy differences between these states appear tovary significantly between kinases. For example, whereas DFG-out conformations seem extraordinarily stable for certain ABLcomplexes, such conformations, although possible, have a highthermodynamic penalty in SRC complexes (43). The reason forthis strikingly differing behavior is unknown and cannot beattributed to a few individual differing amino acids. The role ofsuch residues in the neighborhood of the DFG motif or else-where and the overall energetics of the different states cannotbe determined from the crystal structures alone. NMR-deriveddynamic information, as obtained here, should lead us to betterunderstand these systems and to comprehendwhy certain inac-tive conformations are more or less favorable in some kinasesrelative to others (21).To understand howpointmutations cause patient resistance to

imatinib andeventual relapse is alsoof crucial importancebecausethe efficacy of inhibitors may be related to the free energy land-scape of the various inactive states (43). A comprehensive descrip-tionof this behavior is a fundamental prerequisite for amore ratio-nal design of potent new drugs.

Acknowledgments—We thank Drs. Sonja Alexandra Dames, Sebas-tian Meier, and Martin Allan for help during the initial phase of theproject.

REFERENCES1. Melo, J. V., and Barnes, D. J. (2007) Nat. Rev. Cancer 7, 441–4532. Ren, R. (2005) Nat. Rev. Cancer 5, 172–1833. Huse, M., and Kuriyan, J. (2002) Cell 109, 275–2824. Mol, C. D., Dougan, D. R., Schneider, T. R., Skene, R. J., Kraus, M. L.,

Scheibe, D. N., Snell, G. P., Zou, H., Sang, B. C., and Wilson, K. P. (2004)

J. Biol. Chem. 279, 31655–316635. Griffith, J., Black, J., Faerman, C., Swenson, L.,Wynn,M., Lu, F., Lippke, J.,

and Saxena, K. (2004)Mol. Cell 13, 169–1786. Schindler, T., Bornmann,W., Pellicena, P., Miller,W. T., Clarkson, B., and

Kuriyan, J. (2000) Science 289, 1938–19427. Hubbard, S. R., Wei, L., Elis, L., and Hendrickson, W. A. (1994) Nature

372, 746–7548. Wan, P. T. C., Garnett, M. J., Roe, S. M., Lee, S., Niculescu-Duvaz, D.,

Good, V. M., Jones, C. M., Marshall, C. J., Springer, C. J., Barford, D., andMarais, R. (2004) Cell 116, 855–867

9. Pargellis, C., Tong, L., Churchill, L., Cirillo, P. F., Gilmore, T., Graham,A. G., Grob, P.M., Hickey, E. R.,Moss, N., Pav, S., and Regan, J. (2002)Nat.Struct. Biol. 9, 268–272

10. Druker, B. J., Guilhot, F., O’Brien, S. G., Gathmann, I., Kantarjian, H.,Gattermann, N., Deininger, M.W. N., Silver, R. T., Goldman, J. M., Stone,R.M., Cervantes, F., Hochhaus, A., Powell, B. L., Gabrilove, J. L., Rousselot,P., Reiffers, J., Cornelissen, J. J., Hughes, T., Agis, H., Fischer, T., Verhoef,G., Shepherd, J., Saglio, G., Gratwohl, A., Nielsen, J. L., Radich, J. P., Si-monsson, B., Taylor, K., Baccarani, M., So, C., Letvak, L., and Larson, R. A.(2006) N. Engl. J. Med. 355, 2408–2417

11. Cowan-Jacob, S.W., Fendrich,G., Floersheimer, A., Furet, P., Liebetanz, J.,Rummel, G., Rheinberger, P., Centeleghe, M., Fabbro, D., and Manley,P. W. (2007) Acta Crystallogr. Sect. D Biol. Crystallogr. 63, 80–93

12. Cowan-Jacob, S. W., Guez, V., Fendrich, G., Griffin, J. D., Fabbro, D.,Furet, P., Liebetanz, J., Mestan, J., and Manley, P. W. (2004) Mini-Rev.Med. Chem. 4, 285–299

13. Nagar, B., Hantschel, O., Young, M. A., Scheffzek, K., Veach, D., Born-mann, V., Clarkson, B., Superti-Furga, G., and Kuriyan, J. (2003) Cell 112,859–871

14. Nagar, B., Bornmann, W. G., Pellicena, P., Schindler, T., Veach, D. R.,Miller, W. T., Clarkson, B., and Kuriyan, J. (2002) Cancer Res. 62,4236–4243

15. Manley, P.W., Cowan-Jacob, S.W., Buchdunger, E., Fabbro, D., Fendrich,G., Furet, P., Meyer, T., and Zimmermann, J. (2002) Eur. J. Cancer 38,S19–S27

16. Weisberg, E., Manley, P. W., Breitenstein, W., Bruggen, J., Cowan-Jacob,S. W., Ray, A., Huntly, B., Fabbro, D., Fendrich, G., Hall-Meyers, E., Kung,A. L., Mestan, J., Daley, G. Q., Callahan, L., Catley, L., Cavazzall, C., Azam,M., Neuberg, D., Wright, R. D., Gilliland, G., and Griffin, J. D. (2005)Cancer Cell 7, 129–141

17. Weisberg, E., Manley, P., Mestan, J., Cowan-Jacob, S., Ray, A., and Griffin,J. D. (2006) Br. J. Cancer 94, 1765–1769

18. Das, J., Chen, P., Norris, D., Padmanabha, R., Lin, J., Moquin, R. V., Shen,Z. Q., Cook, L. S., Doweyko, A. M., Pitt, S., Pang, S. H., Shen, D. R., Fang,Q., de Fex, H. F.,McIntyre, K.W., Shuster, D. J., Gillooly, K.M., Behnia, K.,Schieven, G. L., Wityak, J., and Barrish, J. C. (2006) J. Med. Chem. 49,6819–6832

19. Tokarski, J. S., Newitt, J. A., Chang, C. Y. J., Cheng, J. D., Wittekind, M.,Kiefer, S. E., Kish, K., Lee, F. Y. F., Borzillerri, R., Lombardo, L. J., Xie, D. L.,Zhang, Y. Q., and Klei, H. E. (2006) Cancer Res. 66, 5790–5797

20. Verkhivker, G. A. (2007) Biopolymers 85, 333–34821. Wissing, J., Godl, K., Brehmer,D., Blencke, S.,Weber,M.,Habenberger, P.,

Stein-Gerlach, M., Missio, A., Cotten, M., Muller, S., and Daub, H. (2004)Mol. Cell. Proteomics 3, 1181–1193

22. Wisniewski, D., Lambek, C. L., Liu, C. Y., Strife, A., Veach, D. R., Nagar, B.,Young, M. A., Schindler, T., Bornmann, W. G., Bertino, J. R., Kuriyan, J.,and Clarkson, B. (2002) Cancer Res. 62, 4244–4255

23. Kraker, A. J., Hartl, B. G., Amar, A.M., Barvian,M. R., Showalter, H. D. H.,and Moore, C. W. (2000) Biochem. Pharmacol. 60, 885–898

24. Wiesner, S., Wybenga-Groot, L. E., Warner, N., Lin, H., Pawson, T., For-man-Kay, J. D., and Sicheri, F. (2006) EMBO J. 25, 4686–4696

25. Vogtherr, M., Saxena, K., Hoelder, S., Grimme, S., Betz, M., Schieborr, U.,Pescatore, B., Robin, M., Delarbre, L., Langer, T., Wendt, K. U., andSchwalbe, H. (2006) Angew. Chem. Int. Ed. Engl. 45, 993–997

26. Strauss, A., Bitsch, F., Fendrich, G., Graff, P., Knecht, R., Meyhack, B., andJahnke, W. (2005) J. Biomol. NMR 31, 343–349

27. Strauss, A., Bitsch, F., Cutting, B., Fendrich, G., Graff, P., Liebetanz, J.,Zurini, M., and Jahnke, W. (2003) J. Biomol. NMR 26, 367–372




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

28. Cowan-Jacob, S. W., Fendrich, G., Manley, P. W., Jahnke, W., Fabbro, D.,Liebetanz, J., and Meyer, T. (2005) Structure 13, 861–871

29. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A.(1995) J. Biomol. NMR 6, 277–293

30. Johnson, B. A., and Blevins, R. A. (1994) J. Biomol. NMR 4, 603–61431. Ottiger, M., Delaglio, F., and Bax, A. (1998) J. Magn. Reson. 131,

373–37832. Dosset, P., Hus, J. C., Blackledge, M., and Marion, D. (2000) J. Biomol.

NMR 16, 23–2833. Grzesiek, S., and Bax, A. (1993) J. Biomol. NMR 3, 185–20434. Vajpai, N., Strauss, A., Fendrich, G., Cowan-Jacob, S. W., Manley, P. W.,

Jahnke, W., and Grzesiek, S. (2008) Biomol. NMR Assign.10.1007/s12104-008-9079-7

35. Hansen, M. R., Mueller, L., and Pardi, A. (1998) Nat. Struct. Biol. 5,1065–1074

36. Tjandra, N., and Bax, A. (1997) Science 278, 1111–1113

37. Sass, J., Cordier, F., Hoffmann, A., Cousin, A., Omichinski, J. G., Lowen,H., and Grzesiek, S. (1999) J. Am. Chem. Soc. 121, 2047–2055

38. Lipari, G., and Szabo, A. (1982) J. Am. Chem. Soc. 104, 4546–455939. Korzhnev, D. M., Salvatella, X., Vendruscolo, M., Di Nardo, A. A.,

Davidson, A. R., Dobson, C. M., and Kay, L. E. (2004) Nature 430,586–590

40. Young, M. A., Shah, N. P., Chao, L. H., Seeliger, M., Milanov, Z. V., Biggs,W. H., Treiber, D. K., Patel, H. K., Zarrinkar, P. P., Lockhart, D. V. J.,Sawyers, C. L., and Kuriyan, J. (2006) Cancer Res. 66, 1007–1014

41. Levinson, N. M., Kuchment, O., Shen, K., Young, M. A., Koldobskiy, M.,Karplus, M., Cole, P. A., and Kuriyan, J. (2006) PLoS Biol. 4, 753–767

42. Eisenmesser, E. Z., Millet, O., Labeikovsky, W., Korzhnev, D. M., Wolf-Watz,M., Bosco, D. A., Skalicky, J. J., Kay, L. E., andKern, D. (2005)Nature438, 117–121

43. Seeliger, M. A., Nagar, B., Frank, F., Cao, X., Henderson, M. N., andKuriyan, J. (2007) Structure 15, 299–311




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Manley, Stephan Grzesiek and Wolfgang JahnkeNavratna Vajpai, André Strauss, Gabriele Fendrich, Sandra W. Cowan-Jacob, Paul W.

Imatinib/Nilotinib and DasatinibDetermined by NMR Substantiate the Different Binding Modes of

Solution Conformations and Dynamics of ABL Kinase-Inhibitor Complexes

doi: 10.1074/jbc.M801337200 originally published online April 22, 20082008, 283:18292-18302.J. Biol. Chem.

10.1074/jbc.M801337200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

http://www.jbc.org/content/suppl/2008/04/23/M801337200.DC1

http://www.jbc.org/content/283/26/18292.full.html#ref-list-1

This article cites 42 references, 8 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M801337200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;283/26/18292&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/283/26/18292

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=283/26/18292&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/283/26/18292

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/suppl/2008/04/23/M801337200.DC1

http://www.jbc.org/content/283/26/18292.full.html#ref-list-1

http://www.jbc.org/

solutionconformationsanddynamicsofabl kinase ... · solutionconformationsanddynamicsofabl...

Documents