Activity of dual SRC-ABL inhibitors highlights the roleof BCR�ABL kinase dynamics in drug resistanceMohammad Azam*, Valentina Nardi*, William C. Shakespeare†, Chester A. Metcalf III†, Regine S. Bohacek†,Yihan Wang†, Raji Sundaramoorthi†, Piotr Sliz*, Darren R. Veach‡, William G. Bornmann§, Bayard Clarkson‡,David C. Dalgarno†, Tomi K. Sawyer†, and George Q. Daley*¶

*Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, and Division of Hematology�Oncology, The Children’s Hospital,Dana–Farber Cancer Institute, Brigham and Women’s Hospital, Boston, MA 02115; †Ariad Pharmaceuticals, Inc., Cambridge, MA 02139-4234; ‡MemorialSloan–Kettering Cancer Center, New York, NY 10021; and §Department of Experimental Diagnostic Imaging, University of Texas M. D. AndersonCancer Center, Houston, TX 77054

Edited by John Kuriyan, University of California, Berkeley, CA, and approved April 24, 2006 (received for review January 3, 2006)

Mutation in the ABL kinase domain is the principal mechanism ofimatinib resistance in patients with chronic myelogenous leuke-mia. Many mutations favor active kinase conformations that pre-clude imatinib binding. Because the active forms of ABL and SRCresemble one another, we tested two dual SRC-ABL kinase inhib-itors, AP23464 and PD166326, against 58 imatinib-resistant (IMR)BCR�ABL kinase variants. Both compounds potently inhibit mostIMR variants, and in vitro drug selection demonstrates that active(AP23464) and open (PD166326) conformation-specific compoundsare less susceptible to resistance than imatinib. Combinations ofinhibitors suppressed essentially all resistance mutations, with thenotable exception of T315I. Guided by mutagenesis studies andmolecular modeling, we designed a series of AP23464 analogues totarget T315I. The analogue AP23846 inhibited both native andT315I variants of BCR�ABL with submicromolar potency butshowed nonspecific cellular toxicity. Our data illustrate how con-formational dynamics of the ABL kinase accounts for the activity ofdual SRC-ABL inhibitors against IMR-mutants and provides a ratio-nale for combining conformation specific inhibitors to suppressresistance.

kinase inhibitors � imatinib � combination chemotherapy �chronic myelogenous leukemia

The small molecule protein kinase inhibitors imatinib (1–5),gefitinib (6), and erlotinib (7, 8) are susceptible to resistance in

patients because of amino acid substitutions in the target protein.Imatinib inhibits BCR�ABL by stabilizing the kinase in a catalyt-ically inactive conformation (9). Point mutations in the ABL kinasedomain can thwart drug binding by direct steric hindrance or bydestabilizing the inactive kinase conformation that is required forimatinib binding (2, 3). Consequently, developing drugs that targetthe open or active conformation of the kinase may prove effectivein rescuing patients who develop imatinib resistance.

Previously, we carried out an in vitro screen for imatinib resis-tance and identified a large number of mutant amino acid residuesoutside the active site that did not appear to act by direct sterichindrance of drug binding. Several of these residues were homol-ogous to SRC residues known to play critical roles in maintainingan assembled, autoinhibited SRC kinase conformation (10–13),and some previously had been implicated by site-directed mutagen-esis in ABL kinase regulation (14, 15). We reasoned that theseconformational, or allosteric, mutants exerted effects on drugbinding by favoring adoption of the active kinase conformation.Using inferences from the mutagenesis studies, we proposed amodel for the assembled ABL kinase that closely resembled theautoinhibited SRC structure (3). Crystallographic and biochemicaldata published alongside our mutagenesis report confirmed thatABL indeed was regulated in a SRC-like manner (16–18).

The striking similarity between the catalytically active states ofthe SRC and ABL kinases prompted us to investigate whether SRCkinase inhibitors might be effective against imatinib-resistant (IMR)

ABL variants (3, 16). In this report, we have analyzed the activityof AP23464 and PD166326 against 58 IMR variants of BCR�ABLand conducted screens for in vitro resistance to these compoundsindividually and in combination with imatinib. Our data show thatthese agents show potent activity against the majority of IMR

mutants and are less subject to resistance, with the notable excep-tion of T315I. Based on screening and structural analysis, wechemically modified AP23464 to achieve kinase inhibition of T315I,although the compounds suffered from cellular toxicity. Our re-sults, together with structural modeling, provide important insightsinto the role of kinase dynamics in mediating drug resistance andsuggest that a combination of conformation-specific inhibitors caneffectively suppress molecular resistance.

ResultsKinase-Activating IMR BCR�ABL Variants Are Sensitive to AP23464 andPD166326. AP23464 and PD166326 are synthetic small-molecule,ATP-competitive dual-specificity SRC�ABL kinase inhibitors (Fig.1A). AP23464 and PD166326 inhibited the proliferation of BaF3cells transformed with BCR�ABL (BaF3-BCR�ABL) with IC50values of 13.4 and 5.4 nM, respectively. AP23464 and PD166326 areconsiderably more potent than imatinib and showed no inhibitionof untransformed IL-3-dependent BaF3 cells at a concentration of10 �M. To obtain a comparative inhibition profile, we analyzed theactivity of these two compounds against 58 IMR BCR�ABL vari-ants. Both compounds showed greater potency than imatinibagainst most variants, with the notable exception of T315I (Fig. 1B;see also Fig. 5 and Table 1, which are published as supportinginformation on the PNAS web site).

AP23464 was superior to PD166326 at inhibiting the P-loopvariants L248R, G250E, Q252H, Y253H, and E255K. The variantA269V showed significant resistance against AP23464 (IC50 � 201nM; 15-fold greater than for native BCR�ABL) but only modestresistance against PD166326 (IC50 � 19.5 nM; 3.5-fold greater; Fig.1B and Table 1). Interestingly, several variants from the C helix andthe C lobe are more sensitive than native BCR�ABL to AP23464and PD166326 (Fig. 1B and Table 1). We have confirmed thedifferent relative activity profiles of these variants by in vitroautophosphorylation assay (Fig. 5). The higher activity of theAP23464 and PD166326 compounds against the IMR BCR�ABLvariants implies more favorable binding to a distinct conformationalstate promoted by point mutation.

BCR�ABL Mutations Resistant to AP23464. To understand the struc-ture activity relationships and patterns of resistance for the

Conflict of interest statement: No conflicts declared.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviation: IMR, imatinib-resistant.

¶To whom correspondence should be addressed. E-mail: george.daley@childrens.harvard.edu.

© 2006 by The National Academy of Sciences of the USA

9244–9249 � PNAS � June 13, 2006 � vol. 103 � no. 24 www.pnas.org�cgi�doi�10.1073�pnas.0600001103

AP23464 compound, we performed a drug selection screen withmutagenized BCR�ABL, as described for imatinib (3). The yield ofAP23464-resistant colonies was consistently lower than for ima-tinib. At the highest concentration of AP23464 tested (500 nM), theyield of resistant colonies dropped to 3 per 106 cells (Fig. 2A). Thesedata suggest that AP23464 is not only more potent than imatinib butis also less susceptible to resistance.

We sequenced the resistant clones. In 200 nM drug, 20 aminoacid substitutions were discovered affecting 12 different residues.Four of these residues confer imatinib resistance, whereas eight areunique to AP23464. Nine of 13 clones harbored E255K, H295,T315I, F317V, and C330S as a single mutation, whereas threeclones harbored E255K or T315I mutations associated with E373K,D444N, R386S, and L387V (Tables 2 and 3, which are published assupporting information on the PNAS web site). Residue E373 liesat the end of the �7 strand after helix-E, where it appears tomodulate the movement of the N and C lobes relative to each otherthrough direct contact with Y320 and T319 of the hinge region;mutation at E373 likely destabilizes the kinase cleft in a manner thatis unfavorable for AP23464 binding. Residue H295 lies at the Cterminus of helix C and makes contact with S349 of helix E.Mutations at these residues may have a profound effect on the

Fig. 2. Frequency of resistance against AP23464 and PD166326 alone and incombination with imatinib. (A) Frequency of resistance of BAF3 cells trans-formed with randomly mutagenized BCR�ABL. For each experiment, 150 �106 BAF3 cells were transduced with randomly mutagenized BCR�ABL virus.After 16–20 h cells were divided into 15 subgroups and exposed to differentdrug concentrations, as indicated below each bar. Cells were then plated at adensity of 0.25 � 106 per well of a six-well plate. The number of clonesobtained per 106 cells from two independent experiments were averaged andplotted. (B) Frequency of resistance from the BAF3 cells harboring native,nonmutagenized BCR�ABL. For each experiment, 130 � 106 BAF3-BCR�ABLcells were divided into 13 subgroups and exposed to different drug concen-trations, as indicated below each bar. Numbers below each bar representnumber of resistant colonies from 107 BaF3 cells.

Fig. 1. Activity of AP23464, PD166326, and imatinib against IMR BCR�ABLvariants. (A) Structures of AP23464 and PD166326. (B) Fold difference in IC50

values, relative to native BCR�ABL, normalized to 1 and plotted on a semi-logarithmic scale. IMR BCR�ABL variants identified in clinical samples areshown in bold italics. Structural motifs and critical SH3, SH2 contacts areindicated on the left side of the figure.

Azam et al. PNAS � June 13, 2006 � vol. 103 � no. 24 � 9245

MED

ICA

LSC

IEN

CES

kinetics of opening and closing of the catalytic cleft, which maycontribute significant drug resistance. Clones selected in 500 nMAP23464 represented a very limited spectrum of mutations ateither T315I or F317V. One clone harbored both mutations T315Iand F317C. Modeling studies with AP23464 revealed that themutations T315I and F317V�C cause direct steric blockade to drugbinding and alter the hydrophobic surface of the binding site (datanot shown).

BCR�ABL Mutations Resistant to PD166326. We next selected forresistance to PD166326. As for AP23464, selection in PD166326yielded fewer colonies than imatinib (Fig. 2A). Sequence analysisrevealed a range of point mutations identical to IMR variants, andseveral previously undescribed ones (Tables 2 and 3). Like imatinib,mutation in the P loop residues was a predominant mode ofresistance: The mutants G250E and E255K were recovered in 35%and 45% of sequenced clones, respectively. Interestingly, mutationsfrom the Cap, SH3, SH2, CD linker, helices C, E, I, and theactivation loop were frequently found in association with G250E orE255K (Table 3). The Y253H mutant was recovered in 13% ofclones, typically associated with secondary mutations: Q30P (cap),Q252H, A359G, I403L (activation loop), and E281G (C helix).Y128T was identified as a robust single mutation from the SH2domain that conferred resistance to 100 nM PD166326. PD166326-resistant clones that did not have P loop or gatekeeper residuemutations were found to harbor multiple substitutions from distinctkinase regulatory motifs. For instance, one clone harbored fourmutations, some of which previously were linked to imatinibresistance: D233N (CD linker), E292K (C helix), I360L (�-6), andH375P (SH2 contact). Another harbored S385T, A399P, (activa-tion loop), and V422L (helix F, myristate binding pocket). Fourclones recovered at 500 nM PD166326 each harbor T315I substi-tutions, illustrating that this mutation remains refractory to evenhigh concentrations of this potent compound.

Combinations of Inhibitors Select for a Narrow Spectrum of Muta-tions. Although highly potent, the single inhibitors remain suscep-tible to resistance by point mutation in the ABL kinase. PD166326and AP23464 suppress essentially all resistance mutations at 500nM, with the notable exception of T315I and F317V�C. However,it is likely that at this drug concentration, these compounds wouldmanifest undesirable off-target activity against other kinases. Anappealing strategy for suppressing resistance is to combine agents,on the presumption that their spectrum of resistance will notoverlap and that lower concentrations of two agents will suppressresistance. To test this conjecture, we performed in vitro screens forresistance to combinations of the kinase inhibitors at differentsubmaximal concentrations (Fig. 2A and Table 2). Combinations ofAP23464 with PD166326 or imatinib reduced the yield of resistantclones to 3–4 per 106 cells. The resistant clones that survive thecombination of AP23464 with PD166326 harbor T315I and F317Vmutations, whereas clones resistant to AP23464 with imatinibharbor T315I and F317L. The combination of PD166326 withimatinib was subject to a broader spectrum of resistance mutations:Three of four clones harbored E255K mutations, and two showedmutations in the C helix (E281G) or the activation loop (K400Q)and F-helix (E450K) (Table 3). The triple combination of imatinib,PD166326, and AP23464 at 5 �M, 50 nM, and 100 nM, respectively,yielded significantly fewer resistant colonies but failed to suppressE255K and E279K, mutations that are clinically prevalent (Table 4,which is published as supporting information on the PNAS website). Importantly, at higher drug concentrations (200 nM ofAP23464, 100 nM PD166326, and 5 �M of imatinib), resistance wasrare and mediated by the T315I mutation only.

These combination data allow several interesting conclusions: (i)IMR mutations are more apt to be cross-resistant to PD166326 thanto AP23464, (ii) combinations of inhibitors reduce the frequency ofresistance in vitro, and (iii) combining even three compounds at

lower drug concentrations will not suppress P loop variants; instead,high concentrations of each of the drugs is required to be maximallyeffective. These in vitro data suggest that combination therapy maybe an appealing front-line strategy for reducing primary resistance,particularly for the treatment of chronic myelogenous leukemiapatients who have an advanced stage disease at diagnosis and areoften imatinib refractory. Unfortunately, using drugs in combina-tion may not allow for reduced dosing of individual agents, andcombination therapy may not be a means for reducing side effectsof high-dose regimens.

Selection for Drug Resistance Against Native BCR�ABL. In vitro mu-tagenesis of BCR�ABL provides a highly sensitive means ofcataloging resistance mutations that might arise during treatment.To test whether mutagenesis introduces bias into our screen and todocument that mutagenesis enhances screening efficiency, weperformed a screen against cell lines transformed by native, non-mutagenized BCR�ABL. The pattern of mutations that we recov-ered was strikingly similar to screens employing mutagenesis,including the predominance of the P loop mutations Y253H andE255K and the refractory nature of T315I (Fig. 2B; see Tables 5–7,which are published as supporting information on the PNAS website). Interestingly, many of the clones also harbored multiplemutations (Tables 6 and 7), implying that multiple mutations arenot an artifact of hypermutation in our system but are detectedbecause of cooperative effects on drug resistance.

Importantly, the yield of resistant clones from cells trans-formed with nonmutagenized BCR�ABL was �10-fold lowerthan from cells transformed with a randomly mutagenizedlibrary of BCR�ABL plasmids (Fig. 2 A and B). Furthermore,sequence analysis revealed that only 70% of colonies isolated inthe absence of mutagenesis harbored relevant point mutationscompared with 100% in screens from mutagenized libraries,implying that other mechanisms of drug resistance play a largerrole in screens of native BCR�ABL. A previous screen forresistance against PD166326 by using native, nonmutagenizedBCR�ABL captured only a subset of mutations found in ourscreen of mutagenized BCR�ABL and required sequencing ofconsiderably more clones (19). A recently published screenagainst dasatinib (BMS-354825) that used mutagenesis showedefficient identification of resistant clones (20).

AP23464 Variants Designed to Inhibit T315I. A combination ofAP23464 with imatinib suppressed virtually all drug resistance,except for the vexing T315I variant. Structural modeling reveals thatthe phenol ring attached to N-9 of AP23464 targets the deephydrophobic pocket of the ABL kinase active site (Fig. 6 andSupporting Methods, which are published as supporting informationon the PNAS web site); access to that site is limited by the mutantT315I. Therefore, we tested a series of AP23464 analogues withmodifications at the N-9 position for activity against the T315I ABLkinase (Fig. 3A). AP23848 and AP23980 were only weakly effective(IC50 � 6,422 nM and 5,055 nM, respectively), whereas AP23846showed significant potency (IC50 � 297 nM). The key modificationin AP23846 relative to AP23464 (or its exact parent compoundAP23848) is the replacement of the phenol moiety with the ethylgroup at N-9 (Fig. 3A). Replacement of the ethyl group at N-9 withhydrogen (AP23980) resulted in the loss of inhibitory potency.AP23846 can dock within the ABL active site and does not clashwith the T315I substitution (Fig. 3B). AP23846 inhibited theautophosphorylation of both native and T315I-BCR�ABL in acellular lysate at submicromolar concentration (Fig. 3C). Cellulartesting showed that AP23846 inhibited proliferation of BaF3 cellstransformed by native BCR�ABL and the T315I variant; unfortu-nately, it also inhibited proliferation of the parental BAF3 cells,presumably because of off-target effects (Fig. 3D). These findingsdemonstrate that an ATP-competitive inhibitor can be effectiveagainst the T315I mutant by having limited penetration into the

9246 � www.pnas.org�cgi�doi�10.1073�pnas.0600001103 Azam et al.

hydrophobic pocket. However, additional structural features mustbe engineered into our compound to achieve ABL kinase bindingspecificity without undesirable toxicity.

DiscussionMutations in the ABL kinase confer imatinib resistance byeither direct steric hindrance to drug binding or by destabi-lizing the assembled kinase structure that imatinib prefers.The structural resemblance between the catalytically activestates of the SRC and ABL kinases prompted us to test thehypothesis that dual SRC-ABL kinase inhibitors would showactivity against IMR kinase-activating variants. We tested theactivity of two such compounds, AP23464 and PD166326,against a panel of 58 IMR mutants from structurally diverseregulatory and catalytic motifs. Most of the IMR variants aresensitive to one or both of the compounds, with the exceptionof the P loop variants, T315I, and F317L�V�C. The varyingpattern of sensitivity ref lects the different affinities of the twocompounds for the distinct conformations adopted by thevarious mutant forms, highlighting the conformational plas-ticity of the ABL kinase. These data confirm our hypothesisthat regulatory mutants that activate the kinase also becomesusceptible to dual SRC-ABL inhibitors.

The cocrystal structure of PD166326 and ABL shows an openand yet catalytically inactive kinase conformation, with the criticalDFG motif rotated away from the catalytic center. In contrast, thecocrystal of AP23464 and SRC (21) demonstrates a DFG confor-mation that is distinctly more open and catalytically favorable.AP23464 was more effective than PD166326 at suppressing drugresistance, likely because of its association with a more catalyticallyactive kinase conformation. Dasatinib (BMS-354825), which like-wise targets the open conformation of the ABL kinase, also hasshown reduced frequency of resistance mutations in an in vitroscreen for drug resistance (20).

Engineered mutations in the SH2-C-lobe interface and myr-istate-binding pocket release ABL kinase autoinhibition (17), andmutation of this critical regulatory region is a frequent cause ofimatinib resistance (Fig. 4). Interestingly, several of these variantsare hypersensitive to both AP23464 and PD166326, again implyingthat these compounds have higher affinity for the active confor-mation. Cocrystal structures of ABL and PD173955 show the drugcan bind to both the inactive and active conformations, leading tothe concept that dual SRC-ABL inhibitors are ‘‘conformationallytolerant’’ (22, 23). Our data suggests significant variation in theapparent binding affinity for specific BCR�ABL variants, thereby

Fig. 3. AP23846 inhibits T315I-BCR�ABL but shows toxicity in cellular assays.(A) AP23464 analogues with varying modifications at the N-9 position of thepurine template. In this chemical series, the 4-dimethylphosphoryl moiety ofAP23464 was replaced with dipropylphosphoryl to improve oral absorptionand pharmacokinetics, resulting in the parent compound AP23848. Cellularpotency also was enhanced. (B) AP23846 molecule docked onto ABL activesite. Docking was performed by using the 1OPK coordinate and Accelrys DS

MODELING. (C) Immunoblot analysis of the BCR�ABL and T315I-BCR�ABL kinaseactivity against AP23846 at different doses. (D) Cellular proliferation of BAF3,BAF3-BCR�ABL, and BCR�ABL-T315I at different concentrations of AP23846.

Fig. 4. Mutations in a structural module involved in regulating kinasedynamics confer drug resistance. Mutations that affect the SH2-C-lobe inter-face affect the myristate-binding pocket and vice versa. Helix A of the SH2domain is colored yellow; helices E and I of the C lobe are colored in blue andpink, respectively. Amino acid side chains shown in pink are resistant to acombination of imatinib, AP23464, and PD166326; residues in green areresistant to PD166326. Residues that impact directly on myristate binding areshown in solid surface.

Azam et al. PNAS � June 13, 2006 � vol. 103 � no. 24 � 9247

MED

ICA

LSC

IEN

CES

highlighting the plasticity of the ABL kinase and the role of kinasedynamics in mediating drug resistance.

Selection for resistance against AP23464 or PD166326 aloneresulted in the isolation of some previously undescribed mutantsnot associated with imatinib resistance. Previous screens for drugresistance with the compounds dasatinib (20) and PD166326 (24)likewise discovered previously undescribed mutants at residuesG250, E255, T315, and T317, suggesting that each compound willelicit unique variants. Interestingly, identical substitutions of severalP loop (L248, G250, Y253, Q252, and E255) and gatekeeperresidues (T315 and T317) confer cross-resistance to the structurallydistinct compounds imatinib, dasatinib, and PD166326 (20, 24).The common mutants occur chiefly within the active site of theenzyme to which all of the drugs bind. Given the large number ofpossible amino acid substitutions, this observation highlights thestructural constraints imposed on the kinase active site, which mustretain the capacity for ATP binding and hydrolysis to maintaintransforming function. Mutation at residue F317 appears to beparticularly problematic for AP23464 binding because of its abro-gation of Van der Waals interactions with the dimethyl-phosphonylphenyl group of the inhibitor. The reduction in surfacearea provided by the substitution of cysteine or valine residues atF317 destabilizes inhibitor binding. In contrast, the pyridine moietyof imatinib utilizes less surface area from F317 for binding; there-fore, substitution at this position causes only modest resistanceagainst imatinib. The reduced frequency of resistant mutationsagainst dasatinib and AP23464 highlights the importance of tar-geting the active conformation.

Most P loop mutations in our screen carried secondary mutationsin the Cap, SH3, SH2, CD linker, C helix, hinge region, SH2-C-lobeinterface, activation loop, or myristate-binding pocket. Thus, mu-tations in the regulatory domains cooperate with P loop variants toreinforce drug resistance. Such multiple mutations were isolatedeven in screens with nonmutagenized, native BCR�ABL, suggest-ing that multiple mutations are selected functionally for enhanceddrug resistance and do not reflect promiscuous hypermutationduring mutagenesis. We predict that similar combinations of mu-tations will be found in clinical samples. Our data also clearlydemonstrate that exposure of cells to a combination of agents withbinding preferences for different conformations can suppress most,but importantly not all, drug resistance. Even the combination ofPD166326, AP23464, and imatinib at high doses failed to suppressT315I. Combinations at lower doses were not effective and, instead,yielded multiple mutations occurring in concert with E255K andE279K. Neither of these mutations interact directly with the drug,but their association with mutations from regulatory regions (e.g.,the SH2-C-lobe interface and CD linker) suggest that they coop-erate to disrupt the kinase conformation in a manner that precludeseffective binding of all three agents.

ABL Kinase Dynamics Illustrated by Structural Modeling. Mutagenesisfollowed by molecular modeling of drug-resistant variants providescomplementary information to crystallographic data and suggestshypotheses for how mutations might alter the kinase conformation.Mutations affecting the activation loop, helix C, and the SH2-C-lobe interface are very frequent among resistant clones. ThePD173955-ABL structure (1M52) indicates that R386 of the acti-vation loop makes a salt bridge with E292 of the C helix and mayinteract with V289 of the C helix by an induced-dipole moment.Substitution of the polar side group of serine for V289 would predicta stronger contact with the basic side group of R386, which wouldfavor the open ABL conformation (Fig. 7, which is published assupporting information on the PNAS web site). The fact that theIMR S289 variant is hypersensitive to AP23464 and PD166326indicates that the drugs bind preferentially to an open activationloop conformation that may be stabilized by the interaction of R386with V289 and implies a regulatory communication between the Chelix and the activation loop. Such a direct coupling previously has

not been described for the tyrosine kinases, but precedent exists fora comparable interaction in the crystal structure of the Jak3 kinasein an active state (24).

The SH2-C-lobe interface is comprised of helix A from the SH2domain and helices E and I from the C lobe of the kinase. Mutationsin this structural motif presumably disrupt the myristate-bindingpocket or the docking interactions of the SH2-C-lobe interface,both of which are critical to maintaining the ABL kinase in anassembled state (Fig. 8A, which is published as supporting infor-mation on the PNAS web site). Our modeling analysis suggests thatthe E507K mutation may disengage D504 from R170 because of thefavored interaction of D504 with the oppositely charged K507 (Fig.8A). Such an interaction would lead to a straightening of helix-I�,a dislodging of the SH2 domain, and a distortion of the myristatepocket. A change in the orientation of helix E might exert along-range influence on the kinase conformation through disruptedassociations between helices E and C, which also are implied by ourmutagenesis and modeling (Fig. 8 B and C). Residues S349, Y353,and K357 of helix E make direct contact with H295, K294, L298,and E292, respectively, of helix C in the closed conformation, butin the open conformation, the interactions of Y353 and K357 arelost because of the displacement of helices E and C (Fig. 8 A andB). Q346 from helix E makes contact with H375 and V377 from �-8of the activation loop. Our modeling suggests that mutations ofcontact residues from helix E might directly influence the helix Eon helix C movement, thereby altering the kinase conformation ina manner that would alter the active site. The influence of muta-tions affecting the SH2-C-lobe interaction on the conformation ofthe activation loop, and on the movement of helix E and helix C, arehypotheses based on modeling only. Confirmation of any of theabove speculations would require the crystallization of individualmutant forms of BCR�ABL or studies of protein dynamics.

A Proof-of-Concept Compound for Inhibiting T315I. T315 serves as agatekeeper residue that controls access to a hydrophobic region ofthe enzymatic active site that is not contacted by ATP (1, 9).Mutation at this critical residue confers resistance to almost allATP-competitive inhibitors of the ABL kinase (22, 25–32). Acomparable substitution in platelet-derived growth factor receptor(PDGFR) � and PDGFR� can be inhibited by PKC412 andSU6668, respectively, compounds that avoid contact with thehydrophobic region (31, 33). AP23846 is a proof-of-concept com-pound that inhibits the native and T315I kinase and, likewise, avoidsthe hydrophobic pocket (Fig. 6). Unfortunately, AP23846 is not adrug candidate because of nonspecific cytotoxicity, but this com-pound provides incentive to explore other structural modificationsto identify more specific inhibitors of T315I.

Despite remarkable efficacy, imatinib is not a cure for chronicmyelogenous leukemia. Primary refractory disease or relapse withdrug-resistance is particularly problematic in advanced phasechronic myelogenous leukemia (34, 35), and some chronic phasepatients proceed directly to blast crisis despite apparently effectivetreatment with imatinib (36, 37). And although virtually all newlydiagnosed patients attain complete cytogenetic remission afterimatinib, the vast majority retain evidence of residual leukemicclones by PCR (38, 39). The persistence of residual disease raisesthe specter that dormant leukemic clones eventually might escapedrug suppression or provoke disease relapse once imatinib isdiscontinued. Strategies aimed at greater front-line disease eradi-cation and suppression of resistance are needed, most of whichdepend on further research into combination chemotherapy andnext-generation ABL kinase inhibitors that are active against IMR

BCR�ABL variants.

Materials and MethodsCompounds. AP23464 [3-(2-(2-cyclopentyl-6-(4-(dimethylphos-phoryl) phenylamino)-9H-purin-9-yl) ethyl)phenol], AP23848[3-(2-(2-cyclopentyl-6-(4-(dipropylphosphoryl)phenylamino)-

9248 � www.pnas.org�cgi�doi�10.1073�pnas.0600001103 Azam et al.

9H-purin-9-yl)ethyl)phenol], AP23846 [2-cyclopentyl-N-(4-(dimethylphosphoryl)phenyl)-9-ethyl-9H-purin-6-amine], andAP23980 [2-cyclopentyl-N-(4-(dipropylphosphoryl)phenyl)-9H-purin-6-amine] were synthesized at Ariad Pharmaceuticals.PD166326 [6-(2, 6-dichlorophenyl)-2-(3-hydroxymethylphe-nylamino)-8-methyl-8�H�-pyrido [2, 3-�d�] pyrimidine-7-one]was synthesized at the Memorial Sloan–Kettering Cancer Center.

BCR�ABL Mutagenesis and Screening. BCR�ABL mutagenesis andscreening was performed as described in refs. 3 and 40. We chosea graded series of AP23464 concentrations (100, 135, 200, and 500nM), representing 7.5-, 10-, 15-, and 37-fold multiples, respectively,of the IC50 value determined for growth inhibition of BaF3 cellstransformed by native BCR�ABL. The 135 and 200 nM concen-trations of AP23464 correspond to 5 and 10 mM concentrations ofimatinib in the degree of cell inhibition in this assay. Likewise, wechose PD166326 concentrations for screening at 50, 100, and 500nM, where 50 and 100 nM concentrations are comparable with 5and 10 �M imatinib in the inhibitory effect against native BCR�ABL. Genomic DNA was isolated from the resistant clones and aregion of BCR�ABL spanning from the BCR fusion to the Cterminus of the kinase domain (L528; type Ia numbering) was

amplified by PCR with the following primers: BCR�ABL1,5�GAGAACATCCGGGAGCAGCAG�; BCR�ABL2, 5�-CTC-CAGACTGTCCACAGCATTCC-3�; BAKR1, 5�-CTGTCAT-CAACCTGCTCAGGC-3�; and BAKR2, 5�-GCAGCTCTCCT-GGAGGTCCTCG-3�.

Amplified PCR products were sequenced. Sequence alignmentand analysis was performed by DNASTAR II (DNASTAR, Madi-son, WI).

Cell Viability, Kinase Inhibition, and Western Blotting Assay. Cellviability, kinase assay, and Western blotting was performed asdescribed in refs. 3 and 40.

Structural Modeling and Representation. Structural modeling anddocking of the AP23846 on kinase active site were carried out byusing INSIGHT II (Accelrys, Inc., San Diego).

This study was supported by grants from the National Institutes of Health(NIH), the NIH Director’s Pioneer Award of the NIH Roadmap forMedical Research, and the Thomas Anthony Pappas Charitable Foun-dation. G.Q.D. is a recipient of the Burroughs Wellcome Fund ClinicalScientist Award in Translational Research.

1. Gorre, M. E., Mohammed, M., Ellwood, K., Hsu, N., Paquette, R., Rao, P. N.& Sawyers, C. L. (2001) Science 293, 876–880.

2. Shah, N. P., Nicoll, J. M., Nagar, B., Gorre, M. E., Paquette, R. L., Kuriyan,J. & Sawyers, C. L. (2002) Cancer Cell 2, 117–125.

3. Azam, M., Latek, R. R. & Daley, G. Q. (2003) Cell 112, 831–843.4. Deininger, M., Buchdunger, E. & Druker, B. J. (2005) Blood 105, 2640–2653.5. Corbin, A. S., La Rosee, P., Stoffregen, E. P., Druker, B. J. & Deininger, M. W.

(2003) Blood 101, 4611–4614.6. Kobayashi, S., Boggon, T. J., Dayaram, T., Janne, P. A., Kocher, O., Meyerson,

M., Johnson, B. E., Eck, M. J., Tenen, D. G. & Halmos, B. (2005) N. Engl.J. Med. 352, 786–792.

7. Pao, W., Miller, V., Zakowski, M., Doherty, J., Politi, K., Sarkaria, I., Singh,B., Heelan, R., Rusch, V., Fulton, L., et al. (2004) Proc. Natl. Acad. Sci. USA101, 13306–13311.

8. Pao, W., Miller, V. A., Politi, K. A., Riely, G. J., Somwar, R., Zakowski, M. F.,Kris, M. G. & Varmus, H. (2005) PLoS Med. 2, e73.

9. Schindler, T., Bornmann, W., Pellicena, P., Miller, W. T., Clarkson, B. &Kuriyan, J. (2000) Science 289, 1938–1942.

10. Gonfloni, S., Weijland, A., Kretzschmar, J. & Superti-Furga, G. (2000) Nat.Struct. Biol. 7, 281–286.

11. Sicheri, F., Moarefi, I. & Kuriyan, J. (1997) Nature 385, 602–609.12. Xu, W., Harrison, S. C. & Eck, M. J. (1997) Nature 385, 595–602.13. Moarefi, I., LaFevre-Bernt, M., Sicheri, F., Huse, M., Lee, C. H., Kuriyan, J.

& Miller, W. T. (1997) Nature 385, 650–653.14. Brasher, B. B., Roumiantsev, S. & Van Etten, R. A. (2001) Oncogene 20,

7744–7752.15. Pluk, H., Dorey, K. & Superti-Furga, G. (2002) Cell 108, 247–259.16. Nagar, B., Hantschel, O., Young, M. A., Scheffzek, K., Veach, D., Bornmann,

W., Clarkson, B., Superti-Furga, G. & Kuriyan, J. (2003) Cell 112, 859–871.17. Hantschel, O., Nagar, B., Guettler, S., Kretzschmar, J., Dorey, K., Kuriyan, J.

& Superti-Furga, G. (2003) Cell 112, 845–857.18. Harrison, S. C. (2003) Cell 112, 737–740.19. von Bubnoff, N., Barwisch, S., Speicher, M. R., Peschel, C. & Duyster, J. (2005)

Cell Cycle 4, 400–406.20. Burgess, M. R., Skaggs, B. J., Shah, N. P., Lee, F. Y. & Sawyers, C. L. (2005)

Proc. Natl. Acad. Sci. USA 102, 3395–3400.21. Dalgarno, D., Stehle, T., Narula, S., Schelling, P., van Schravendijk, M. R.,

Adams, S., Andrade, L., Keats, J., Ram, M., Jin, L., et al. (2006) Chem. Biol.Drug Des. 67, 46–57.

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

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

24. Boggon, T. J., Li, Y., Manley, P. W. & Eck, M. J. (2005) Blood 106, 996–1002.25. Dorsey, J. F., Jove, R., Kraker, A. J. & Wu, J. (2000) Cancer Res. 60, 3127–3131.26. von Bubnoff, N., Veach, D. R., Miller, W. T., Li, W., Sanger, J., Peschel, C.,

Bornmann, W. G., Clarkson, B. & Duyster, J. (2003) Cancer Res. 63, 6395–6404.27. Shah, N. P., Tran, C., Lee, F. Y., Chen, P., Norris, D. & Sawyers, C. L. (2004)

Science 305, 399–401.28. Weisberg, E., Manley, P. W., Breitenstein, W., Bruggen, J., Cowan-Jacob,

S. W., Ray, A., Huntly, B., Fabbro, D., Fendrich, G., Hall-Meyers, E., et al.(2005) Cancer Cell 7, 129–141.

29. Fabian, M. A., Biggs, W. H., III, Treiber, D. K., Atteridge, C. E., Azimioara,M. D., Benedetti, M. G., Carter, T. A., Ciceri, P., Edeen, P. T., Floyd, M., etal. (2005) Nat. Biotechnol. 23, 329–336.

30. Carter, T. A., Wodicka, L. M., Shah, N. P., Velasco, A. M., Fabian, M. A.,Treiber, D. K., Milanov, Z. V., Atteridge, C. E., Biggs, W. H., III, Edeen, P. T.,et al. (2005) Proc. Natl. Acad. Sci. USA 102, 11011–11016.

31. Blencke, S., Zech, B., Engkvist, O., Greff, Z., Orfi, L., Horvath, Z., Keri, G.,Ullrich, A. & Daub, H. (2004) Chem. Biol. 11, 691–701.

32. O’Hare, T., Pollock, R., Stoffregen, E. P., Keats, J. A., Abdullah, O. M.,Moseson, E. M., Rivera, V. M., Tang, H., Metcalf, C. A., III, Bohacek, R. S.,et al. (2004) Blood 104, 2532–2539.

33. Cools, J., Stover, E. H., Boulton, C. L., Gotlib, J., Legare, R. D., Amaral, S. M.,Curley, D. P., Duclos, N., Rowan, R., Kutok, J. L., et al. (2003) Cancer Cell 3,459–469.

34. Druker, B. J., Sawyers, C. L., Kantarjian, H., Resta, D. J., Reese, S. F., Ford,J. M., Capdeville, R. & Talpaz, M. (2001) N. Engl. J. Med. 344, 1038–1042.

35. Sawyers, C. L., Hochhaus, A., Feldman, E., Goldman, J. M., Miller, C. B.,Ottmann, O. G., Schiffer, C. A., Talpaz, M., Guilhot, F., Deininger, M. W., etal. (2002) Blood 99, 3530–3539.

36. Avery, S., Nadal, E., Marin, D., Olavarria, E., Kaeda, J., Vulliamy, T., BritoBabapulle, F., Goldman, J. M. & Apperley, J. F. (2004) Leuk. Res. 28, Suppl.1, S75–S77.

37. Xu, Y., Wahner, A. E. & Nguyen, P. L. (2004) Arch. Pathol. Lab. Med. 128,980–985.

38. Faderl, S., Hochhaus, A. & Hughes, T. (2004) Hematol. Oncol. Clin. North Am.18, 657–670, ix–x.

39. Paschka, P., Merx, K. & Hochhaus, A. (2004) Acta Haematol. 112, 85–92.40. Azam, M., Raz, T., Nardi, V., Opitz, S. L. & Daley, G. Q. (2003) Biol. Proced.

Online 5, 204–210.

Azam et al. PNAS � June 13, 2006 � vol. 103 � no. 24 � 9249

MED

ICA

LSC

IEN

CES