Discovery and SAR of 1,3,4-thiadiazole derivatives as potent Abl tyrosine kinase inhibitors and cytodifferentiating agents

Article abstract of DOI:10.1016/j.bmcl.2007.11.112

A series of substituted benzoylamino-2-[(4-benzyl)thio]-1,3,4-thiadiazoles has been discovered as potent Abl tyrosine kinase inhibitors. Molecular docking simulations on the Abl tyrosine kinase were conducted in order to rationalize the SAR of the synthesized inhibitors. The most active compound identified from the enzymatic screening (6a) showed interesting inhibitory activity on Imatinib-sensitive murine myeloid 3B clone and Bcr-Abl-independent Imatinib-resistant leukemia cells. Surprisingly, 6a was also proved to act as differentiating inducers in human promyelocytic leukemia cells (HL-60).

Full text of DOI:10.1016/j.bmcl.2007.11.112

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1207–1211
Discovery and SAR of 1,3,4-thiadiazole derivatives as potent
Abl tyrosine kinase inhibitors and cytodifferentiating agents
Marco Radi,a Emmanuele Crespan,b Giorgia Botta,a Federico Falchi,a Giovanni Maga,b
Fabrizio Manetti,a Valentina Corradi,a Manuela Mancini,c
Maria Alessandra Santucci,c Silvia Schenoned and Maurizio Bottaa,*
a
`
Dipartimento Farmaco Chimico Tecnologico, Universita degli Studi di Siena, Via Alcide de Gasperi 2, I-53100 Siena, Italy
bIstituto di Genetica Molecolare, IGM-CNR, Via Abbiategrasso 207, I-27100 Pavia, Italy
c
`
Istituto di Ematologia e Oncologia Medica ‘‘Lorenzo e Ariosto Seragnoli’’, University of Bologna-Medical School,
Via Massarenti 9, 40138 Bologna, Italy
dDipartimento di Scienze Farmaceutiche, University of Genoa, V.le Benedetto XV, 3 Genoa, Italy
Received 15 November 2007; revised 27 November 2007; accepted 29 November 2007
Available online 4 December 2007
Abstract—A series of substituted benzoylamino-2-[(4-benzyl)thio]-1,3,4-thiadiazoles has been discovered as potent Abl tyrosine
kinase inhibitors. Molecular docking simulations on the Abl tyrosine kinase were conducted in order to rationalize the SAR of
the synthesized inhibitors. The most active compound identified from the enzymatic screening (6a) showed interesting inhibitory
activity on Imatinib-sensitive murine myeloid 3B clone and Bcr-Abl-independent Imatinib-resistant leukemia cells. Surprisingly,
6a was also proved to act as differentiating inducers in human promyelocytic leukemia cells (HL-60).
Ó 2007 Elsevier Ltd. All rights reserved.
O
Targeted therapies have become one of the latest inno-
N
N
N
N
O
HN
vative trends in the treatment of cancer, trying to inhibit
a specific molecular target that is believed to have a crit-
ical role in tumor growth or progression.1 The FDA’s
approval of Imatinib (GleevecTM, STI571, Fig. 1) for
the treatment of chronic myeloid leukemia (CML)
serves as validation of the concept that therapeutic
agents targeting cancer-specific pathways can offer signi-
ficative improvements over traditional chemotherapic
agents.2 The efficacy of GleevecTM can be attributed to
the critical role of Abl TK constitutive activation by
Bcr in the disease pathogenesis and to the drug specific-
ity as an Abl inhibitor.3 However, the initial enthusiasm
generated by the high response rate to this drug has been
dampened by the development of resistance, accounting
for 16% in newly diagnosed chronic phase CML and
more than 50% in more advanced stages.4 The two main
mechanisms of resistance are the overexpression of Bcr-
Abl, mainly due to gene amplification,5 and, more fre-
quently, mutation in the kinase domain of Bcr-Abl.6
N
S
H
1
N
HN
Ki (μM): Src/Abl = 2.9/NA
N
N
GleevecTM
O
N
N
N
S
2
H
H3C
I
N
K
i (μM): Src/Abl = NA/0.4
N
N
O
Cl
N
O
N
H
N
N
S
N
OH
NH
CH3
N
S
SprycelTM
H
3
I
K
i (μM): Src/Abl = 2.0/1.3
Figure 1. FDA-approved TK inhibitors (GleevecTM, SprycelTM) and
selected thiadiazoles (1–3).
In addition, point-mutations in the Bcr-Abl kinase do-
main do not account for all cases of resistance. Blast cri-
sis is in fact usually associated with additional genetic
changes conferring Bcr-Abl-independent pathways of
proliferation and maturation arrest. The greatest limit
of Gleevec arises from its ability to bind an unique, inac-
tive conformation of the Abl TK in which the activation
Keywords: Tyrosine kinase inhibitors; 1,3,4-Thiadiazoles; Bcr-Abl;
Leukemia; SAR.
*
Corresponding author. E-mail: botta@unisi.it
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.112

1208
M. Radi et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1207–1211
loop is in a closed position.7 This characteristic reduces
or abrogates drug effectiveness when Abl TK is stabi-
lized in its active conformation in consequence of auto-
phosphorylation events or mutations. An important step
forward has been made with the recent approval of
Dasatinib (SprycelTM, Fig. 1), a multiply targeted tyro-
sine kinase inhibitor, active against 14 of the 15 clini-
cally relevant Gleevec resistant mutants.8 Acquired
resistance is becoming a common issue in targeted ther-
apy of malignant diseases and, as a consequence, there is
a growing interest in developing novel TK inhibitors
able to target Imatinib-resistant form of CML.9
latter compounds was split in three portions and reacted
with three different benzoyl chlorides (in the presence of
the same scavenger) thus obtaining, after removal of the
resin, a small library of 21 compounds (6a–u). These fi-
nal compounds were then screened, in a cell-free assay,
against recombinant human c-Src and Abl using Imati-
nib as a reference compound (Table 1). Interestingly,
several of the tested derivatives were characterized by
high affinity toward the isolated Abl with an inhibitory
activity (expressed as Ki values) comparable to that of
Imatinib. A significative improvement has been there-
fore reached with respect to the parent compounds 1–
3. In addition, the metabolic stability of compounds
6a–u in the presence of CYP3A4 was modeled using
molecular interaction field descriptors calculated in Vol-
Surf.14 As a result, the thiadiazole derivatives showed a
metabolic stability comparable to or, sometimes, even
better than Imatinib (Fig. 2).
A computational protocol, based on docking/MD simu-
lations and a pharmacophore-based database search,
has been recently developed by our research group lead-
ing to the identification of commercially available com-
pounds showing moderate inhibitory activity toward
both Src and Abl kinases.10 Among these new molecular
scaffolds (original in the field of both Src and Abl inhib-
itors), the thiadiazole derivatives 1–3 attracted our
attention due to their ability to inhibit both Src and
Abl (dual inhibitors) or alternatively only one of the
two kinases, depending on the substitution pattern on
the benzamido moiety (Fig. 1). To fully understand
the SAR of these new Src/Abl inhibitors, we planned
to synthesize a small library of thiadiazoles (analogues
of 1–3) bearing a wide range of substituents on both aro-
matic rings. Herein we describe the synthesis, anticancer
activity, and docking studies of novel benzoylamino-2-
[(4-benzyl)thio]-1,3,4-thiadiazoles 6a–u. For the most
active compound (6a), a preliminary study on its cyto-
differentiating properties on HL-60 cells is also de-
scribed.11 A two-step parallel solution phase approach
has been used to generate the combinatorial library of
compounds 6a–u (Scheme 1).
The substitution pattern on the two aromatic rings of
6a–u plays therefore an important role in determining
the affinity for the enzyme Abl. However, while it was
evident that the less active compounds 6r–u were charac-
terized by the presence of an ortho-chloro substituent on
the benzamide moiety, the rationalization of the SAR
for the remaining derivatives required the intervention
of 3D considerations. Accordingly, after validation of
the computational protocol,16 compounds 6a–u were
docked into the ATP binding site of Abl17 both in its
inactive conformation (PDB code: 1IEP, cocrystallized
with Imatinib)18 and active conformation (PDB code:
Table 1. c-Src/Abl tyrosine kinase inhibitory activities exerted by
compounds 6a–u
Compound
R1
R2
Activitya (Ki, lM)
c-Src
Abl
In the first step, the commercially available 5-amino-
1,3,4-thiadiazole-2-thiol (4) was partitioned into seven
separate reaction vessels and then reacted with seven dif-
ferent benzyl bromides using the resin AmberliteÒ IRA-
67 both as a base and as a scavenger for the hydrobro-
mic acid released in the reaction medium as a side prod-
uct.12 Once all the reactions were completed (24 h), the
resin was removed by parallel filtration to give the inter-
mediates 5a–g in more than 95% purity.13 Each of the
6a
6b
6c
p-F
p-F
0.354
0.217
0.464
0.195
0.219
0.221
0.165
0.200
0.569
0.199
0.263
0.170
0.064
0.522
0.718
0.247
0.334
0.900
0.169
1.137
0.272
31
0.044
0.047
0.070
0.073
0.083
0.089
0.092
0.104
0.167
0.189
0.195
0.210
0.217
0.225
0.272
0.369
0.400
0.406
0.760
0.920
1.260
0.013
p-Br
H
p-Cl
p-F
6d
6e
m-Cl
p-NO2
p-Br
p-NO2
p-F
p-Cl
o-Cl
p-F
6f
6g
6h
6i
p-Cl
p-Cl
p-F
m-F
6j
p-OCH3
p-OCH3
p-NO2
m-F
p-Cl
p-F
6k
6l
p-F
R1
6m
6n
6o
6p
6q
6r
p-Cl
o-Cl
o-Cl
p-F
N
N
i.
N
N
p-Br
m-F
S
H2N
SH
H2N
S
S
m-Cl
H
5a-g
4
p-Cl
o-Cl
o-Cl
o-Cl
o-Cl
p-F
ii.
6s
m-Cl
H
R1
6t
O
R2
N
N
6u
Imatinib
p-OCH3
S
N
S
H
a Values are means of at least two experiments and were calculated
according to the following equation: Ki = ID50/{E0 + [E0 (Km(ATP)/
S0)]}/E0, where E0 and S0 are the enzyme and the ATP concentration
(0.125 and 0.160 pmol, respectively, for Src; 0.022 and 0.016 lM,
respectively, for Abl).
6a-u
Scheme 1. Reagents and conditions: (i) benzyl bromides, AmberliteÒ
IRA-67, CH3CN, 24 h, rt; (ii) benzoyl chlorides, AmberliteÒ IRA-67,
THF, 24 h, 60 °C.

M. Radi et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1207–1211
1209
the HY I in order to allow the amide NH group to estab-
lish a hydrogen bond with Thr315. However, the loss of
both a hydrogen bond contact with Met318 and impor-
tant van der Waals interactions with the HY II resulted
in a reduced affinity for the enzyme (Fig. 3b). An inter-
esting exception was represented by those compounds
bearing a p-NO2 substituent on the benzylthio moiety
(Fig. 3c). In the case of compound 6e, the loss of inter-
actions with both Met318 and HY II was compensated
by the formation of three hydrogen bonds, one with
Asp381 and two more with Asn322. For the remaining
compounds, an intermediate binding mode between
the two above described gave account for the low-micro-
molar activity values. Based on the docking study, it can
be therefore speculated that compounds 6a–u may inter-
act preferably with the Abl kinase in its active
conformation.
Figure 2. Projection of CYP3A4 metabolic stability PLS score for
compounds 6a–u.
The effect of 6a on the reproductive integrity of clonal
myeloid progenitors (32D) stably transduced with Bcr-
Abl either sensitive or resistant to Gleevec was then eval-
uated. Results shown in Figure 4 prove that compound
6a significantly reduces the clonogenic activity
(LD50 = 2.2 lM) of a Bcr-Abl-expressing 32D clone sen-
2GQG, cocrystallized with Dasatinib).15 While no signi-
ficative information was obtained from the docking
studies on the inactive form (all the docked compounds
assumed the same binding mode),19 a nice 3D-SAR was
found from the docking analysis on the active confor-
mation of Abl. Alternatively, two different binding
modes were found for the synthesized compounds: the
most active inhibitors 6a–d were characterized by a lin-
ear binding mode (similar to that of Dasatinib) with the
thiadiazole nucleus located in the region usually occu-
pied by the thiazole ring of Dasatinib. The amide moiety
of 6a–d was oriented outwards in order to give, together
with the thiadiazole nitrogen, crucial hydrogen bond
interactions with Met318 (Fig. 3a). The two phenyl rings
were accommodated within two distinct hydrophobic re-
gions (HY I and HY II, Fig. 3) located at opposite sides
of the thiadiazole ring; several hydrophobic contacts
were found with residues in the HY I (namely, Val256,
Lys271, Met290, Val299, and Ile313) and HY II
(namely, Leu248, Gly249, and Gly321). On the other
hand, the less active compounds 6r–u were characterized
by a folded binding mode (Fig. 3b): the presence of a
ortho-chloro substituent seems to induce a reorientation
of the molecule with the benzamide moiety occupying
Figure 4. Clonogenic assay. Drug sensitivity of clone 3B, subclone 4.2,
and clone 3B + IL-3 to Imatinib (IM) and compound 6a.
Figure 3. Docked conformations of representative thiadiazoles into the ATP binding site of the Dasatinib:Abl complex (PDB code: 2GQG).15 (a)
Docked conformation of 6a (sticks, purple carbon atoms); (b) docked conformation of 6u (sticks, yellow carbon atoms); (c) docked conformation of
6e (sticks, green carbon atoms). Dashed lines, hydrogen bonding. HY I (hydrophobic region I), HY II (hydrophobic region II).

1210
M. Radi et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1207–1211
Figure 5. Antiproliferative (A), cytotoxic (B), and cytodifferentiating (C) effect of 6a on PMA or zymosan stimulated induced HL-60 cells. AUC
(area under the curve); CL index (chemiluminescence index).
sitive to Gleevec (3B, Fig. 4) and of a subclone where
drug resistance was driven by a growth factor (IL-3)
autocrine loop (4.2, Fig. 4). Moreover, compound 6a
was effective against Bcr-Abl-expressing 32D cells ren-
dered Gleevec insensitive by IL-3 saturating doses
(3B + IL-3, Fig. 4). The effects of compound 6a on
p210Bcr-Abl phosphorylation were investigated in clone
3B at 33 °C and clone 3B IM resistant at 24 h of expo-
sure to 5 and 10 lM of the drug (Fig. 6). Our results
demonstrate that the exposure to 6a 10 lM sensibly
abrogates p210Bcr-Abl phosphorylation at Tyr245, both
in clone 3B IM sensible and in clone 3B IM resistant.
Notably, p210 Bcr-Abl TK inhibition by 6a was associ-
ated with a significant decrease of clonogenic activity.
All results concerning p210Bcr-Abl phosphorylation sig-
nals evoked by 6a were repeated three times.
PMA or zymosan (Fig. 5C). The antiproliferative activ-
ity of 6a on HL-60 cells can be therefore related to a
cytodifferentiating effect.
In summary, a small library of novel benzoylamino-2-
[(4-benzyl)thio]-1,3,4-thiadiazoles (6a–u) has been
quickly synthesized using a parallel solution phase ap-
proach. Some of these compounds were characterized
by high affinity toward the isolated Abl with an inhibi-
tory activity (expressed as Ki values) comparable to that
of Imatinib. Docking studies, conducted to understand
the SAR, suggested a preferential interaction with the
active conformation of Abl. The identified hit com-
pound 6a showed a promising biological profile, being
able to target both Bcr-Abl-related (clone 3B, Imatinib
sensitive) and Bcr-Abl-independent leukemia cells (sub-
clone 4.2 and clone 3B+IL-3, Imatinib resistant). In
addition, compound 6a was also proved to act as an
interesting cytodifferentiating agent on leukemia cells
(HL-60) characterized by a different pathogenetic path-
way. Further studies are ongoing in our laboratories.
Those results indicate that the thiadiazoles herein de-
scribed may circumvent CML progenitor resistance to
TK inhibitors arising from autophosphorylation of
Bcr-Abl protein.20 Finally, the effect of compound 6a
on a tumor cell line (HL-60) characterized by a different
pathogenetic pathway was also evaluated. As shown in
Figure 5A, compound 6a showed significant dose-
dependent inhibitory effects on the growth rate of HL-
60 cells, cultured for 96 h. Surprisingly, this inhibitory
effect did not seem to depend on a direct cytotoxic ac-
tion of the inhibitors, being the viability indexes (used
to define the percentage of cell viability) unmodified
up to 100 lM of 6a (Fig. 5B). The restoration of so-
called respiratory burst was evaluated as fundamental
functional marker of differentiation in human leukemia
cell lines.21 This aspect of phagocyte oxidative metabo-
lism has been analyzed by chemiluminescence (CL) in
HL-60 cells, incubated with different concentrations of
6a for 96 h, and stimulated by phorbol-12-myristate-
13-acetate (PMA) and zymosan. Results clearly showed
a dose-dependent recovery of reactive oxygen species
(ROS) metabolism when the cells were stimulated by
Acknowledgments
The University of Siena and Fondazione Monte dei
Paschi di Siena are greatly acknowledged for financial
support. Dr. Giovanni Gaviraghi (SienaBiotech) is also
acknowledged for helpful discussions. One of us (GB)
thanks Prof. Bruno Giardina and Prof. Roberto Scatena
`
(Universita Cattolica del Sacro Cuore, Rome) for host-
ing her in their laboratory. We also acknowledge Prof.
Gabriele Cruciani (University of Perugia, Italy) for
kindly providing us with the program VolSurf.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2007.11.112.
References and notes
1. Sawyers, C. Nature 2004, 432, 294.
2. Schindler, T.; Bornmann, W.; Pellicena, P.; Miller, W. T.;
Clarkson, B.; Kuriyan, J. Science 2000, 289, 1938.
3. Druker, B. J. Adv. Cancer Res. 2004, 91, 1.
Figure 6. Effects of 6a on the phosphorylation of p210Bcr-Abl
.

M. Radi et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1207–1211
1211
4. Druker, B. J. et al. N. Engl. J. Med. 2006, 356, 2408.
5. Hochhaus, A.; Kreil, S.; Corbin, A. S.; La Rosee, P.;
Muller, M. C.; Lahaye, T.; Hanfstein, B.; Schoch, C.;
Cross, N. C.; Berger, U.; Gschaidmeier, H.; Druker, B. J.;
Hehlmann, R. Leukemia 2002, 16, 2190.
6. Gorre, M. E.; Mohammed, M.; Ellwood, K.; Hsu, N.;
Paquette, R.; Rao, P. N.; Sawyers, C. L. Science 2001, 293,
876.
7. Azam, M.; Latek, R. R.; Daley, J. Q. Cell 2003, 112, 831.
8. Tokarski, J. S.; Newitt, J. A.; Chang, C. Y.; Cheng, J. D.;
Wittekind, M.; Kiefer, S. E.; Kish, K.; Lee, F. Y. F.;
Borzillerri, R.; Lombardo, L. J.; Xie, D.; Zhang, Y.; Klei,
H. E. Cancer Res. 2006, 66, 5790.
14. (a) Cruciani, G.; Pastor, M.; Guba, W. Eur. J. Pharm. Sci.
2000, 11, S29; (b) Cruciani, G.; Crivori, P.; Carrupt, P.-A.;
Testa, B. J. Mol. Struct. 2000, 503, 17.
15. Tokarski, J. S.; Newitt, J.; Chang, C. Y. J.; Cheng, J. D.;
Wittekind, M.; Kiefer, S. E.; Kish, K.; Lee, F. Y. F.;
Borzilerri, R.; Lombardo, L. J.; Xie, D.; Zhang, Y.; Klei,
H. E. Cancer Res. 2006, 66, 5790.
16. Details on the computational protocol are reported in the
Supporting information.
17. Steady-state kinetic studies conducted on 6a showed
that this compound is a competitive inhibitor with
respect to ATP (for details, see Supporting
information).
9. (a) Tauchi, T.; Ohyashiki, K. Int. J. Hematol. 2006, 83,
294; (b) Buschbeck, M. Drugs R&D 2006, 7, 73; (c) Cortes,
J. Curr. Opin. Hematol. 2006, 13, 79.
18. Nagar, B.; Bornmann, W.; Pellicena, P.; Schindler, T.;
Veach, D. R.; Miller, W. T.; Clarkson, B.; Kuriyan, J.
Cancer Res. 2002, 62, 4236.
10. Manetti, F.; Locatelli, G. A.; Maga, G.; Schenone, S.;
Modugno, M.; Forli, S.; Corelli, F.; Botta, M. J. Med.
Chem. 2006, 49, 3278.
11. Botta, M.; Radi, M.; Manetti, F.; Botta, G.; Maga, G.;
Borlak, J. RM2007A000222 (Italian Patent).
19. Data not shown.
20. Mancini, M.; Brusa, G.; Zuffa, E.; Corrado, P.; Martinelli,
G.; Grafone, T.; Barbieri, E.; Cantucci, M. A. Leuk. Res.
2007, 31, 979.
21. Mai, A.; Massa, S.; Rotili, D.; Pezzi, R.; Bottoni, P.;
Scatena, R.; Meranerd, J.; Broschd, G. Bioorg. Med.
Chem. Lett. 2005, 15, 4656.
12. Gayo, L. M.; Suto, M. J. Tetrahedron Lett. 1997, 38, 513.
13. Determined via HPLC/MS, see Supporting information.