Design and synthesis of thiadiazoles and thiazoles targeting the Bcr-Abl T315I mutant: From docking false positives to ATP-noncompetitive inhibitors

Article abstract of DOI:10.1002/cmdc.201000066

(Figure Presented) Overcoming resistance: In an effort to optimize our previously identified dual Src/Abl hits, a new series of 1,3,4-thiadiazoles and 1,3-thiazoles were designed and synthesized, paying particular attention to the reduction of their lipophilicity and to the improvement of the affinity towards the drug-resistant T315I mutant. Compound 5 was identified as a promising allosteric inhibitor of the T315I mutant.

Full text of DOI:10.1002/cmdc.201000066

MED
DOI: 10.1002/cmdc.201000066
Design and Synthesis of Thiadiazoles and Thiazoles Targeting the Bcr-Abl
T315I Mutant: from Docking False Positives to ATP-Noncompetitive
Inhibitors
Marco Radi,[a] Emmanuele Crespan,[b] Federico Falchi,[a] Vincenzo Bernardo,[a] Samantha Zanoli,[b]
Fabrizio Manetti,[a] Silvia Schenone,[c] Giovanni Maga,[b] and Maurizio Botta*[a, d]
Chronic myeloid leukemia (CML) was the first neoplastic dis-
ease for which the knowledge of the molecular pathogenesis
led to the development of a curative therapy. Imatinib mesy-
late (Gleevec; Novartis, Basel, Switzerland), the first targeted
drug for the treatment of CML, has a high rate of remission
and is currently used as frontline therapy.[1] However, patients
treated with imatinib (IM) can develop resistance to the drug
leading to leukemia progression.[2] Multiple mechanisms of re-
sistance have been identified, though the dominant mecha-
nism seems to be represented by amino acid point mutations
within the kinase domain of Bcr-Abl.
(3) are rare examples of ATP-competitive inhibitors that are
also able to inhibit those enzymes with the T315I mutation.[6]
Second generation Bcr-Abl inhibitors (such as dasatinib and
nilotinib) are capable of inhibiting many IM-resistant forms of
the kinase but not the form in which threonine is mutated to
isoleucine at the highly conserved gatekeeper residue Thr315
(T315I). This specific mutation accounts for 15% of all point
mutations and confers complete resistance to all ATP-competi-
tive Bcr-Abl inhibitors currently on the market.[3] The mutation
of the gatekeeper residue in Bcr-Abl (Thr315) to a bulkier hy-
drophobic residue (such as isoleucine) often causes the loss of
an important hydrogen bond necessary for high-affinity inhibi-
tor binding and creates steric hindrance that could interfere
with inhibitor binding within the ATP-binding pocket.[4] In addi-
tion, recent studies showed that bulkier hydrophobic residues
in position 315 tend to stabilize a “hydrophobic spine&-
rdquo;, shifting the equilibrium towards the active conforma-
tion of Bcr-Abl.[5] For this reason, targeting Bcr-Abl with com-
pounds able to avoid unfavorable clashes with Ile315 is be-
coming an effective strategy in the search for new drugs active
on the highly resistant T315I mutant. The aurora kinase/Abl in-
hibitors VX-680 (1), PHA-739358 (2) and the Abl inhibitor PPY-A
The pyrazolo[3,4-d]pyrimidines (such as 4), synthesized by
our research group, have recently shown promising inhibitory
activity against the T315I mutant enzyme and have also exhib-
ited potent cytotoxicity to cells isolated from IM-resistant pa-
tients.[7] Another recently explored strategy to overcome T315I
drug resistance involves the development of small molecules
targeting enzyme regions outside the ATP binding site (ATP-
noncompetitive inhibitors) leading to agents that should be
unaffected by mutations in the kinase domain that cause re-
sistance to the common ATP-competitive inhibitors (e.g., imati-
nib, nilotinib and dasatinib).[8] A few ATP-noncompetitive inhib-
itors have already been identified (e.g., GNF2,[9] ON012380,[10]
DCC-2036[11]) and proven to be unaffected by the Bcr-Abl T315I
mutation.
[a] Dr. M. Radi, Dr. F. Falchi, Dr. V. Bernardo, Dr. F. Manetti, Prof. M. Botta
Dipartimento Farmaco Chimico Tecnologico, University of Siena
Via Alcide de Gasperi 2, 53100 Siena (Italy)
Fax: (+39)0577-234333
E-mail: botta.maurizio@gmail.com
[b] Dr. E. Crespan, Dr. S. Zanoli, Dr. G. Maga
Istituto di Genetica Molecolare, IGM-CNR
Via Abbiategrasso 207, 27100 Pavia (Italy)
[c] Prof. S. Schenone
Dipartimento di Scienza Farmaceutiche, Universitꢀ degli Studi di Genova
Viale benedetto XV 3, 16132 Genova (Italy)
In the search for new molecular scaffolds active on Bcr-Abl,
our research group has recently identified a family of 1,3,4-
thiadiazole derivatives as interesting inhibitors of Bcr-Abl (e.g.,
compounds 5 and 6).[12] The lead compound 5 showed inter-
esting inhibitory activity on murine myeloid clones transduced
with IM-sensitive or resistant Bcr-Abl. Kinetic studies showed
[d] Prof. M. Botta
Sbarro Institute for Cancer Research and Molecular Medicine, Center for
Biotechnology, College of Science and Technology, Temple University, BioLife
Science Bldg., Suite 333, 1900 N 12th Street, Philadelphia, PA 19122 (USA)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201000066.
1226
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1226 – 1231

that compound 5 acts as a Bcr-Abl ATP-competitive inhibitor,
and the subsequent docking simulations on the active form of
the enzyme allowed structure–activity relationships to be sug-
gested for this family of compounds.[12]
rameters affecting drug-likeness and efficiency.[14] With this
concept in mind we decided to maintain the 4-F- and 4-NO2
benzylthio chain in the 5-position of the thiadiazole and to in-
troduce an additional nitrogen on the solvent-exposed benz-
amido moiety (10a and 11 a,b; Scheme 1). In addition, the su-
perimposition of the leads 5 and 6 within the Bcr-Abl binding
site (figure S1, Supporting Information) suggests that the intro-
duction of a hydrogen-bond acceptor (a carbonyl moiety) be-
tween the sulfur and the substituted phenyl ring could give
additional hydrogen-bonding interactions with the hinge
region for those compounds bearing a 4-nitrobenzylthio
moiety in C2-position of the thiadiazole (11 e; Scheme 1; see
also figure S1c in the Supporting Information).
Following a previously reported protocol,[12] the commercial-
ly available 5-amino-1,3,4-thiadiazole-2-thiol (8) was reacted
with the selected alkylating agents in the presence of the resin
Amberliteꢁ IRA-67 (acting both as a base and as an acid scav-
enger) to give the intermediates 9a–e in high yield and purity.
The final compound 10a was then obtained by coupling 9a
with 3-amino-4-chlorobenzoic acid in the presence of EDC and
HOBt and purified by simple trituration in dichloromethane
and petroleum ether. To obtain compounds 11 a–e, the inter-
mediates 9a–e were dissolved in anhydrous THF and treated
with nicotinoyl chloride hydrochloride in the presence of an
excess of pyridine. The final products 11 a–e were precipitated
by addition of water to the reaction mixtures and isolated as
pure compounds via filtration (Scheme 1).
In the proposed binding mode, both compounds 5 and 6
failed to establish any contacts with the gatekeeper residue
Thr315 (Figure 1a and b) and could therefore represent an in-
teresting starting point for the design of novel analogues tar-
geting the T315I Bcr-Abl mutant. Subsequent molecular mod-
eling studies suggested that the nitrogen atom in the 4-posi-
tion of the thiadiazole ring may not be necessary for interac-
tion with Met318 and that the phenyl ring of the benzamido
moiety could be profitably functionalized. The thiazole deriva-
tives 7 showed an improved affinity towards wild-type (wt)
Bcr-Abl.[13] However, the common drawback of the hits 5–7
was represented by their high lipophilicity. Accordingly, the
present study has been focused on the synthesis of new 1,3,4-
thiadiazole and thiazole analogues (general structure I,
Figure 1) endowed with a less lipophilic character and poten-
tially active on the T315I mutant.
In the first part of this work, we focused on the synthesis of
1,3,4-thiadiazole analogues endowed with a predicted solubili-
ty profile better than that of hits 5 and 6 (set 1). The highest
probability of reducing lipophilicity of a series of compounds
can be obtained by replacing lipophilic carbon atoms for nitro-
gen atoms thus reducing LogD with minimal impact on molec-
ular mass. As recently reported, the latter two parameters are
the most important factors in determining the permeability of
a drug candidate and are also connected with a variety of pa-
Taking into account the hit-to-lead optimization process that
led to the discovery of Dasatinib, we also decided to introduce
Figure 1. Docking poses of compounds a) 5 and b) 6 within
the ATP binding site of Bcr-Abl (PDB code: 2GQG). Hydro-
gen bonds with key residues are represented as dashed
black lines. c) The generic structure (I) of the target com-
pounds.
ChemMedChem 2010, 5, 1226 – 1231
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1227

MED
and 33a,b). The common precursor 5-[(4-nitrophe-
nyl)methyl]-2-thiazolamine (22) was synthesized fol-
lowing the Meerwein protocol reported by Obushak
et al. (Scheme 3).[16] Commercially available 4-NO2-ani-
line (19) was converted into the corresponding diazo-
nium salt 20, which was directly added to a toluene
solution of acrolein in the presence of a cupric salt
(CuCl2·2H2O) as the catalyst to obtain the unstable a-
chloro aldehyde 21. This intermediate was treated,
without further purification, with thiourea in refluxing
ethanol to give the desired thiazole derivative 22 as
a brown solid. A first set of final compounds (23a–d)
was then obtained by acylation of 22 with a series of
Scheme 1. Reagents and conditions: a) R1Cl or R1Br, Amberliteꢁ IRA-67, CH3CN, RT, 24 h
(for 9a,b); or THF, reflux, 48 h (for 9c); or DMF, RT, 1.5 h (for 9d,e); b) EDC, HOBT, DMAP,
DMF, RT, overnight; c) nicotinoyl chloride hydrochloride, THF, pyridine, 608C, 48 h.
a piperazinyl ethanol moiety (characteristic of Dasatinib) on
the solvent-exposed benzamido phenyl ring of hit 6, trying to
improve its solubility profile. The target compound 18 was
synthesized starting from commercial 4-F-benzoic acid 12,
which was initially converted into the corresponding ethyl
ester 13 and then submitted to a nucleophilic substitution
with hydroxyethyl piperazine[15] to give compound 14. Protec-
tion of the free hydroxy group of 14 with TBDMSCl and subse-
quent deprotection of the ester moiety gave the carboxylic
acid 16. Coupling of this product with 5-(4-nitrobenzylthio)-
1,3,4-thiadiazol-2-amine (9b) and deprotection with TBAF gave
the desired product 18 (Scheme 2).
In the second part of this work, we focused on the synthesis
of thiazole analogues specifically designed to target the ATP
binding pocket of the T315I mutant (set 2). A cross docking
study was conducted on the three available crystal structures
of the Bcr-Abl enzyme bearing the T315I mutation (PDB: 2Z60,
2V7A, 3K7) to design new analogues potentially endowed with
a specificity for the gatekeeper mutant (See Supporting Infor-
mation, figure S2). This study prompted us to synthesize a
family of thiazole analogues characterized by a 4-NO2-benzyl
group in the 5-position of the thiazole scaffold (general struc-
ture I; X=CH, Y=CH2) and bearing different functional groups
bound to the 2-position of the same scaffold (23a–d, 29a,b
Scheme 3. Reagents and conditions: a) NaNO2, HCl (3n), 08C, 40 min; b) acro-
lein, CuCl2·2H2O, toluene, 3 h, RT; c) thiourea, EtOH, 5 d, reflux; d) Ar-
(CH2)nCOCl, Amberliteꢁ IRA-67, THF, 24 h, 608C.
selected acyl chlorides following a previously reported proto-
col.[12]
To prepare 29a,b, we initially envisioned the regioselective
opening of a substituted styrene epoxide by reaction with 22
in the presence of indium tribromide (InBr3)[17] to give
a b-amino alcohol that could have been oxidized to
give the desired compounds (Scheme 4). The 2-(3-
methoxyphenyl)oxirane (24), prepared by oxidation
of 3-vinylanisole with mCPBA,[18] was reacted with 22
in the presence of InBr3 to give the undesired b-
amino alcohol 25 resulting from the epoxide opening
via the endo-N of 22. The structure of compound 25
was further confirmed by NOE experiments. Although
certain electrophiles (e.g., alkyl halides) are known to
react at the endo-N of thiazoles, others (such as acid
chlorides and anhydrides) react preferentially at the
exo-N.[19] Accordingly, compound 22 was initially pro-
tected at the exo-N (compound 26, Scheme 4) and
subsequently treated with the substituted 2-bromo-
acetophenones 27a,b to give a mixture of endo-N-
Scheme 2. Reagents and conditions: a) EtOH, H2SO4 (cat), 24 h, 808C; b) 1-(2-hydroxy-
ethyl)piperazine, CH3CN, 60 h, 808C; c) TBDMSCl, imidazole, DMF, 15 min, RT; d) NaOH 1m,
CH3OH, 30 min, 658C; e) 9b, POCl3, DMAP, toluene, 15 min, 1308C; f) TBAF, DMF, 96 h, RT.
and exo-N-alkyl products in a 1:1 ratio. These deriva-
tives (28a,b) were isolated by chromatographic pu-
1228
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1226 – 1231

center of the Golden Triangle
showing a marked improvement
with respect to the lead 5. On
the other hand, compounds be-
longing to set 2 were designed
with the sole intention of im-
proving the affinity of the agent
for the T315I mutant and, un-
fortunately, were predicted to
show a behavior similar to or
worse than that of the lead 5.
According to the Golden Trian-
gle predictions, it would there-
fore be easier to identify poten-
Scheme 4. Reagents and conditions: a) InBr3, CH2Cl2, RT, 24 h; b) (Boc)2O, DMAP, THF, 1 h, 608C; c) NaH, DMF, 1 h,
08C then 24 h, RT; d) HCl/AcOEt (1:1), 24 h, RT.
rification and deprotected to give the final compounds 29a,b.
Table 1. Enzymatic activity and selected molecular descriptors of the
The thiazole derivatives 33a,b were prepared from the com-
mercially available acetophenones. Wittig homologation of
compounds 30a,b gave the aldehydes 32a,b, which were fi-
nally reacted with the 2-aminothiazole 22 under reductive ami-
nation conditions to give the desired final products 33a,b
(Scheme 5).
target compounds.
Entry
Compd
MW
cLogD[a]
Ki[b] [mm]
Abl wt
Src
Abl T315I
1
2
5 (BO1)
6
363.4
406.9
330.4
394.9
346.4
373.4
329.4
374.4
401.4
500.6
353.4
383.4
373.8
345.4
387.8
383.4
387.9
383.5
3.92
4.36
3.81
3.56
2.81
2.55
1.39
2.47
2.16
2.19
3.74
3.71
4.43
3.63
4.37
3.96
6.17
5.57
0.55
0.17c
ND
NA
NA
0.10
0.10[c]
0.02[c]
1.40
1.22
0.20
NA
0.41
0.20
2.95
0.90
1.30
0.80
0.80
0.55
0.60
0.36
1.20
0.40
ND
ND
NA
NA
17.20
NA
13.90
15.20
ND
28.14
ND
16.00
28.80
NA
3
7
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
10a
11 a
11 b
11 c
11 d
11 e
18
23a
23b
23c
23d
29a
29b
33a
33b
NA
2.50
0.47
NA
NA
1.13
3.00
1.30
1.11
3.10
2.00
0.31
NA
NA
18.50
NA
[a] Predicted lipophilicity; values calculated at pH 7.4 using the web ver-
sion of the ToxBoxes software.[21] [b] Values are the mean of at least two
experiments. [c] These activity values were obtained from previous ex-
periments as reported in References [12,13].
Scheme 5. Reagents and conditions: a) (Ph)3PCH2OMeCl, nBuLi (1.6m in THF),
THF, 24 h; b) HBr (48%), acetone/H2O (4:1), RT, 2 h; c) 22, NaBH(OAc)3, AcOH,
DME, RT, 72 h.
tial drug candidates among the compounds belonging to
set 1.
As recently reported by Johnson and co-workers, molecular
weight (MW) and liphophilicity (LogD at pH 7.4) act as surro-
gates of many different molecular descriptors and have been
used to develop a useful visualization tool—the Golden Trian-
gle.[20] According to this tool, drug candidates with good per-
meability and low clearance (drug-like space) should be posi-
tioned close to the center of the Golden Triangle (LogD=1.5;
MW=350). In order to obtain some preliminary insight into
the drug-likeness of the synthesized compounds, the LogD
values of the final compounds and the lead 5 were calculated
using the web version of the ToxBoxes software[21] (see Table 1)
and were plotted against the corresponding molecular weights
(Figure 2). As expected, the compounds synthesized to de-
crease the lipophilicity (set 1) are located close to or within the
All final compounds were then screened in a cell-free assay
against recombinant human c-Src, Abl and T315I Abl using 5
as a reference compound (Table 1). Unexpectedly, most of the
derivatives belonging to set 1 (Table 1, entries 4–10) complete-
ly lost their activity towards Src, despite their high structural
similarity to the hit compounds 5 and 6. However, in the same
subset of compounds, 11 c,d represent exceptions (Table 1, en-
tries 7–8); while 11 c was found to be active towards Src only,
compound 11 d behaved as a dual Src/Abl inhibitor (as does 5)
and showed also an appreciable activity against the mutant
T315I. The most interesting compounds 11 b,e were complete-
ly inactive against Src and maintained the activity of the lead 5
towards Abl (wt) while showing an appreciable activity against
ChemMedChem 2010, 5, 1226 – 1231
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1229

MED
Table 2. Inhibition of wild-type (wt) Abl and T315I Abl by compound 5.
Abl
Ki[a] [mm]
Variable substrate
Type of inhibition
wt
wt
T315I
T315I
0.1Æ0.01
0.12Æ0.01
0.4Æ0.1
ATP
peptide
ATP
competitive
mixed
noncompetitive
noncompetitive
0.45Æ0.1
peptide
[a] Values are the mean of at least two experiments. Further details can
be found in the Supporting Information.
the inhibitor AP24534 (PDB code: 3K3).[23] Unfortunately, dock-
ing simulations conducted on this site were unable to rational-
ize the activity of our inhibitors against the T315I mutant. An-
other known allosteric pocket of Bcr-Abl is represented by the
myristate binding site.[8] A co-crystal structure of Bcr-Abl (PDB
code: 3K5V) bound by an ATP-competitive inhibitor (IM) and
an allosteric inhibitor (GNF-2) has been recently released.[24]
This represents the first structure of an allosteric inhibitor
bound to the myristate pocket.
Figure 2. The Golden Triangle. Compounds of set 1 are depicted as black
squares, compounds of set 2 are depicted as grey circles, reference com-
pound 5 is depicted as black triangle.
Docking simulations on the myristate binding site showed
that, among the synthesized compounds, only compound 5
showed a profitable interaction pattern within the pocket. The
4-fluorobenzylthio chain completely fill the hydrophobic
tunnel defined by the residues Leu448, Phe512, Ala363,
Ile451, Ile521, Ala356, Leu360 and Val487, while an hydrogen
bond with Ala452 and edge-to-face aromatic interactions with
Tyr452 contribute to maintain compound 5 in a binding orien-
tation similar to that of the original ligand GNF-2 (Figure 3).
The other synthesized compounds, especially those character-
ized by a polar 4-nitrophenyl substituent, showed less profita-
ble interactions with the myristate pocket, fitting only partially
within the hydrophobic tunnel and remaining mostly solvent
the T315I mutant. Unfortunately, the docking simulations were
not able to explain the loss of activity against Src observed for
most of the compounds belonging to set 1. On the other
hand, the thiazole derivatives suggested by the cross docking
study (set 2) and specifically designed to target the T315I mu-
tants (Table 1, entries 11–18) were found to be much less
active against the mutant than predicted and generally acted
as dual Src/Abl inhibitors.
A thorough kinetic study has been conducted on the lead
compound 5.[22] Surprisingly, this study showed that 5 acts on
the wild-type and mutated T315I Abl enzyme with two differ-
ent mechanisms-of-action (Table 2): 1) In the case of wt Abl,
compound 5 targeted both the
free enzyme and the enzyme–
peptide complex, preventing
ATP binding with a competitive
or mixed mechanism-of-action;
2) in the case of T315I Abl, com-
pound 5 acted only as a non-
competitive inhibitor with re-
spect to both ATP and the pep-
tide substrates. These results
may explain the overestimated
activity for the set 2 compounds,
which were designed for the op-
timal interaction with the ATP
binding pocket of the mutated
T315I Abl but that could interact
with a different allosteric site in
the presence of such mutation.
We initially speculated that
the T315I mutation could force
our compounds to bind in the
deep allosteric pocket on the
Figure 3. Superimposition of the energetically preferred docked conformation of 5 (sticks, gray carbons) and the
crystallized inhibitor GNF-2 (sticks, white carbons) within the myristate binding pocket of Bcr-Abl (PDB code:
back of the gatekeeper residue,
which is partially exploited by 3K5V). Intermolecular hydrogen bonds are highlighted by black dashed lines.
1230
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1226 – 1231

[1] R. Hehlmann, A. Hochhaus, M. Baccarani, Lancet 2007, 370, 342–350.
[2] a) J. F. Apperley, Lancet Oncol. 2007, 8, 1018–1029; b) J. F. Apperley,
Lancet Oncol. 2007, 8, 1116–1128.
exposed. These findings, although speculative in nature, may
support the kinetic studies conducted on the lead compound
5 and suggest the myristate pocket as a possible binding site.
Further experimental data are, however, needed to confirm
this hypothesis.
[3] S. Storey, Nat. Rev. Drug Discovery 2009, 8, 447–448.
[4] A. Hochhaus, S. Kreil, A. S. Corbin, P. La Rosee, M. C. Muller, T. Lahaye, B.
Hanfstein, C. Schoch, N. C. Cross, U. Berger, H. Gschaidmeier, B. J.
Druker, R. Hehlmann, Leukemia 2002, 16, 2190–2196.
The work described here is focused on the hit optimization
of previously identified dual Src/Abl inhibitors based on 1,3,4-
thiadiazole and thiazole scaffolds. Two different set of com-
pounds were designed to solve the two major issues: the high
lipophilicity of the hits and their low affinity for the T315I
mutant. A few interesting inhibitors, endowed with better pre-
dicted drug-like properties, were identified among the com-
pounds of set 1. Compounds 11 c and 11 e, with significant af-
finity for Src and Abl, respectively, will be further explored to
better understand the molecular determinants required for the
inhibition of these two enzymes.
[5] a) A. P. Kornev, N. M. Haste, S. S. Taylor, L. F. T. Eyck, Proc. Natl. Acad. Sci.
USA 2006, 103, 17783–17788; b) M. Azam, M. A. Seeliger, N. S. Gray, J.
Kuriyan, G. Q. Daley, Nat. Struct. Mol. Biol. 2008, 15, 1109–1118.
[6] a) T. Zhou, L. Parillon, F. Li, Y. Wang, J. Keats, S. Lamore, Q. Xu, W. Shake-
speare, D. Dalgarno, X. Zhu, Chem. Biol. Drug Des. 2007, 70, 171;
b) M. A. Young, N. P. Shah, L. H. Chao, M. Seeliger, Z. V. Milanov, W. H.
Biggs III, D. K. Treiber, H. K. Patel, P. P. Zarrinkar, D. J. Lockhart, C. L. Saw-
yers, J. Kuriyan, Cancer Res. 2006, 66, 1007–1014; c) M. Modugno, E.
Casale, C. Soncini, P. Rosettani, R. Colombo, R. Lupi, L. Rusconi, D. Fan-
celli, P. Carpinelli, A. D. Cameron, A. Isacchi, J. Moll, Cancer Res. 2007, 67,
7987–7990.
[7] a) M. A. Santucci, V. Corradi, M. Mancini, F. Manetti, M. Radi, S. Sche-
none, M. Botta, ChemMedChem 2009, 4, 118–126; b) M. A. Santucci, M.
Mancini, V. Corradi, I. Iacobucci, G. Martinelli, M. Botta, S. Schenone,
Invest. New Drugs 2009; DOI: 10.1007/s10637-009-9294-9.
[8] a) M. A. Bogoyevitch, D. P. Fairlie, Drug Discovery Today 2007, 12, 622–
633; b) L. O. Kirkland, C. McInnes, Biochem. Pharmacol. 2009, 77, 1561–
1571.
[9] F. J. Adriꢂn, Q. Ding, T. Sim, A. Velentza, C. Sloan, Y. Liu, G. Zhang, W.
Hur, S. Ding, P. Manley, J. Mestan, D. Fabbro, N. S. Gray, Nat. Chem. Biol.
2006, 2, 95–102.
[10] K. Gumireddy, S. J. Baker, S. C. Cosenza, P. John, A. D. Kang, K. A. Robell,
M. V. Reddy, E. P. Reddy, Proc. Natl. Acad. Sci. USA 2005, 102, 1992–
1997.
[11] R. A. Van Etten, W. W. Chan, V. M. Zaleskas, P. Evangelista, K. Lazarides, C.
Peng, S. Li, S. C. Wise, P. Petillo, D. L. Flynn, Blood (ASH Annual Meeting
Abstracts) 2007, 110, 463.
[12] M. Radi, E. Crespan, G. Botta, F. Falchi, G. Maga, F. Manetti, V. Corradi, M.
Mancini, M. A. Santucci, S. Schenone, M. Botta, Bioorg. Med. Chem. Lett.
2008, 18, 1207–1211.
Surprisingly, the compounds designed to target the T315I
mutants (set 2) showed an overestimated predicted activity
against this mutant and a dual Src/Abl profile comparable to
that of the hit compound 5. A kinetic enzymatic study con-
ducted on the lead 5 showed that this compound was able to
inhibit the drug-resistant T315I Abl mutant by adapting its
mechanism-of-action to the specific enzymatic form of Abl. At
the same time, compound 5 was shown to be an ATP-competi-
tive inhibitor of wt Abl (targeting the free enzyme) and a
purely noncompetitive-ATP inhibitor in the case of Abl T315I
(targeting all enzymatic forms). In an attempt to speculate on
a possible binding site for our compounds, docking simula-
tions suggested the myristate binding pocket as a possible site
that could be target by compound 5 in the presence of the
T315I mutation. Besides providing an explanation for the false
positive docking results on the designed T315I inhibitors,
these data pave the way for the synthesis of novel allosteric in-
hibitors of the Abl T315I mutant. Further studies are ongoing
in our laboratories and will be reported in due course.
[13] F. Manetti, F. Falchi, E. Crespan, S. Schenone, G. Maga, M. Botta, Bioorg.
Med. Chem. Lett. 2008, 18, 4328–4331.
[14] M. J. Waring, Bioorg. Med. Chem. Lett. 2009, 19, 2844–2851.
[15] F. Nobs, (Novartis, AG., Basel, Switzerland), WO/2007/025696, 2007.
[16] N. D. Obushak, V. S. Matiichuk, R. Ya. Vasylyshin, Yu. V. Ostapyuk, Russ. J.
Org. Chem. 2004, 40, 383–389.
[17] J. R. Rodriguez, A. Navarro, Tetrahedron Lett. 2004, 45, 7495–7498.
[18] Data not shown.
[19] R. Barone, M. Chanon, R. Gallo, in Thiazole and Its Derivatives, Part 2,
(Ed.: J. V. Metzger), Wiley, New York, 1979, Chapter 6.
Experimental Section
[20] T. W. Johnson, K. R. Dress, M. Edwards, Bioorg. Med. Chem. Lett. 2009,
19, 5560–5564.
[21] Pharma Algorithms 1, ADME/Tox WEB, http://www.pharma-algorithms.
com/webboxes/ (Last accessed: May 4, 2010).
[22] E. Crespan, M. Radi, S. Zanoli, S. Schenone, M. Botta, G. Maga, Bioorg.
Med. Chem. 2010; DOI: 10.1016/j.bmc.2010.04.024.
Representative and general experimental procedures, compound
characterization data (1H and 13C NMR, elemental analysis), details
of the enzymatic assays and modeling simulations are available in
the Supporting Information.
[23] T. O’Hare, W. C. Shakespeare, X. Zhu, C. A. Eide, V. M. Rivera, F. Wang,
L. T. Adrian, T. Zhou, W. S. Huang, Q. Xu, C. A. Metcalf III, J. W. Tyner,
M. M. Loriaux, A. S. Corbin, S. Wardwell, Y. Ning, J. A. Keats, Y. Wang, R.
Sundaramoorthi, M. Thomas, D. Zhou, J. Snodgrass, L. Commodore, T. K.
Sawyer, D. C. Dalgarno, M. W. Deininger, B. J. Druker, T. Clackson, Cancer
Cell 2009, 16, 401–412.
[24] J. Zhang, F. J. Adriꢂn, W. Jahnke, S. W. Cowan-Jacob, A. G. Li, R. E. Iacob,
T. Sim, J. Powers, C. Dierks, F. Sun, G. R. Guo, Q. Ding, B. Okram, Y. Choi,
A. Wojciechowski, X. Deng, G. Liu, G. Fendrich, A. Strauss, N. Vajpai, S.
Grzesiek, T. Tuntland, Y. Liu, B. Bursulaya, M. Azam, P. W. Manley, J. R.
Engen, G. Q. Daley, M. Warmuth, N. S. Gray, Nature 2010, 463, 501–506.
Acknowledgements
This work was partially supported by an Italian Association for
Cancer Research (AIRC) grant IG 2008–2010 to GM and by a Na-
tional Interest Research Project (PRIN_2007_N7KYCY) to GM and
MB. Fondazione Monte dei Paschi di Siena is also acknowledged
for financial support. SZ was supported by a Fondazione Buzzati-
Traverso Fellowship. EC is the recipient of a FIRC Fellowship
2008–2010. This research has been developed under the umbrella
of CM0602 COST Action.
Received: February 15, 2010
Revised: April 29, 2010
Keywords: Bcr-Abl · chronic myeloid leukemia · T315I ·
thiadiazoles · thiazoles
Published online on May 27, 2010
ChemMedChem 2010, 5, 1226 – 1231
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1231