Structure-based optimization of tyrosine kinase inhibitor CLM3. design, synthesis, functional evaluation, and molecular modeling studies.

Article abstract of DOI:10.1021/jm401358b

Recent advances in the knowledge of thyroid carcinomas development identified receptor tyrosine kinases, like VEGFR2 and RET, as viable and promising targets. Accordingly, their inhibition is emerging as the major therapeutic strategy to treat these pathologies. In this study we describe the synthesis and the functional evaluation of three different series of 4-substituted pyrazolo[3,4-d]pyrimidine derivatives, 8a-g, 9a-g, and 10a-g, designed exploiting a structure-based optimization of the previously developed inhibitor CLM3. Compared to the lead, the novel compounds markedly improved both their inhibitory profile against the target proteins, VEGFR2 and RET, and their antiproliferative efficacy against the medullary thyroid cancer cell line TT. Significantly, compounds 8b, 9c, and 10c proved to block the kinase activity of the mutant RET^V804L, which still lacks effective inhibitors.

Full text of DOI:10.1021/jm401358b

Article
pubs.acs.org/jmc
Structure-Based Optimization of Tyrosine Kinase Inhibitor CLM3.
Design, Synthesis, Functional Evaluation, and Molecular Modeling
Studies.
Stefania Sartini,† Vito Coviello,† Agostino Bruno,‡ Valeria La Pietra,‡ Luciana Marinelli,‡
Francesca Simorini,† Sabrina Taliani,† Silvia Salerno,† Anna Maria Marini,† Anna Fioravanti,§
Paola Orlandi,§ Alessandro Antonelli,§ Federico Da Settimo,† Ettore Novellino,‡ Guido Bocci,§,∥
and Concettina La Motta*,†
†Dipartimento di Farmacia, Universita
̀
̀
di Pisa, Via Bonanno 6, 56126 Pisa, Italy
‡Dipartimento di Farmacia, Universita
di Napoli “Federico II”, Via D. Montesano, 49, 80131 Napoli, Italy
§Dipartimento di Medicina Clinica e Sperimentale, Universita
∥Istituto Toscano Tumori, Via T. Alderotti 26N, 50139 Firenze, Italy
̀
di Pisa, Via Roma 55, 56126 Pisa, Italy
S
* Supporting Information
ABSTRACT: Recent advances in the knowledge of thyroid carcinomas
development identified receptor tyrosine kinases, like VEGFR2 and RET,
as viable and promising targets. Accordingly, their inhibition is emerging
as the major therapeutic strategy to treat these pathologies. In this study
we describe the synthesis and the functional evaluation of three different
series of 4-substituted pyrazolo[3,4-d]pyrimidine derivatives, 8a−g, 9a−g,
and 10a−g, designed exploiting a structure-based optimization of the
previously developed inhibitor CLM3. Compared to the lead, the novel
compounds markedly improved both their inhibitory profile against the
target proteins, VEGFR2 and RET, and their antiproliferative efficacy
against the medullary thyroid cancer cell line TT. Significantly,
compounds 8b, 9c, and 10c proved to block the kinase activity of the
mutant RETV804L, which still lacks effective inhibitors.
the knowledge of pathogenic mechanisms leading to TCs
INTRODUCTION
■
clearly demonstrate that oncogenic tyrosine kinases sustain
their development and/or progression, thus identifying these
proteins as new and promising targets. “Gain of function”
mutations and/or rearrangement of tyrosine kinase receptors
like EGFR and RET, as well as signaling molecules acting
downstream through BRAF, RAS/RAF/ERK, and RAS/PI3K/
AKT pathways, can trigger cell transformations in TCs, thus
providing mitogenic and survival signals. Mutations of the
tyrosine kinase receptor RET have been identified in about 98%
of inherited medullary thyroid carcinoma (MTC) cases at the
germ-line level and in 30−50% of the sporadic cases at the
somatic level. RET rearrangement is also the most common
genetic abnormality in papillary thyroid carcinoma (PTC)
induced by radiation. EGFR and its ligands are implicated in
TCs, having been found in all types of TCs. Moreover, TCs
often exhibit markedly increased vascularisation and elevated
VEGF expression. Indeed, VEGF receptors have been found on
TC cell lines and tumor endothelial cells, and increased VEGF
levels have been determined in the serum/plasma of thyroid
Thyroid carcinomas (TCs) are the most common endocrine
tumors, accounting for up to 1% of all human malignancies.
Their worldwide incidence rate has been increasing sharply
since the mid-1990s, thus becoming the fastest increasing type
of cancer in both men and women in Western countries.
According to American Cancer Society forecasts, in 2013 there
will be about 60 000 novel cases of TCs diagnosed in U.S., with
three to four cases occurring in women, and almost 2000 deaths
due to TCs in both sexes, thus indicating a marked acceleration
in the incidence rate of the past decade of 5.5% and 6.6% per
year in men and women, respectively.1−3
To date, the treatment of choice for different forms of TCs
relies in their complete surgical resection, possibly associated
with external radiation therapy. However, disease can persist or
recur with local and distant metastases, and in these cases,
outcomes are poor, with 5-year survival rates lower than 50%,
often associated with a marked resistance to cytotoxic
chemotherapy.
Indeed, endocrine cancers are generally resistance to DNA-
damaging chemotherapies or radiotherapy, normally exploited
to lead cancer cells to apoptosis.4,5 Accordingly, novel
therapeutic strategies are urgently needed. Recent advances in
Received: August 2, 2013
Published: January 21, 2014
© 2014 American Chemical Society
1225
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
Chart 1. Tyrosine Kinase Inhibitors
cancer patients. These outcomes correlate positively with high
tumorigenic potential, thus clearly highlighting angiogenesis as
an important factor in tumor growth and metastasis. Protein
kinases therefore represent a sound target for the molecular
therapy of people affected by TCs, and their inhibition is now
accepted as an effective therapeutic strategy.6−12
Different approaches have been considered to block the
kinase activity of the mentioned proteins, the most pursued one
being the use of small molecules able to compete with the ATP
site of their catalytic domain. These include compounds already
approved throughout the world for the treatment of other types
of tumors, like sorafenib13 and sunitinib14 (Chart 1), and novel
derivatives such as ponatinib,15 nintedanib,16 axitinib,17 and
pazopanib18 (Chart 1), either proposed or already enrolled in
phases II and III clinical trials for TCs. However, to date, the
only commercially available tyrosine kinase inhibitors approved
by FDA for the treatment of symptomatic or progressive
advanced and metastatic MTC are the quinazolin-4-amino
derivative vandetanib19 and the quinoline derivative cabozanti-
nib20 (Chart 1). Similar to the previously mentioned
derivatives, they both are multitarget compounds, being able
to combine a potent antiangiogenic activity, mainly due to
VEGFR2 inhibition, with inhibitory properties against addi-
tional protein kinases. Accordingly, affecting simultaneously
different mitogenic pathway in both cancer cells and
vasculature, they are able to control both tumor growth and
1226
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
Scheme 1. Synthesis of Pyrazolo[3,4-d]pyrimidine Derivatives 8a−g, 9a−g, and 10a−g
metastasis. However, their significant efficacy is often associated
with serious toxicities, including osteonecrosis, fatal hemor-
rhages, hypertension, QTc prolongation, and respiratory
failures.21 Moreover, as regard to vandetanib, acquired
resistance has been indicated, mainly driven by point mutations
of the target proteins, which strongly limits its exploitation as a
plain and widespread therapeutic agent against TCs.22
Accordingly, the search of novel tyrosine kinase inhibitors is
highly welcome, as novel clinically effective and safe
compounds to treat TCs are strongly required.
Our research group has been involved in the development of
novel tyrosine kinase inhibitors for many years, to achieve novel
drug candidates for the treatment of TCs. Focusing on the
pyrazolo[3,4-d]pyrimidine heterocyclic core, a privileged
scaffold in the tyrosine kinase inhibitors field,23−26 we recently
disclosed a new derivative, (R)-1-phenethyl-N-(1-phenylethyl)-
1H-pyrazolo[3,4-d]pyrimidin-4-amine, namely, CLM3 (Chart
1), which proved to combine an excellent antiangiogenic
efficacy with a less marked inhibitory activity against human
thyroid cancer cell lines, both demonstrated through time- and
concentration-dependent antiproliferative and proapoptotic
activities on specific cell lines.27,28 Actually, the microvascular
endothelial cell line HMVEC-d was remarkably sensitive to low
concentrations of CLM3 (IC50 = 0.40 0.22 nM after 72 h of
exposure), whereas thyroid cancer cell lines as the undiffer-
entiated 8305C (IC50 = 9.20 5.06 μM after 72 h of exposure),
obtained at reoperation from patients with recurrence of the
tumor, CLM3 proved to inhibit significantly its proliferation,
increasing the percentage of apoptotic cells and inhibiting their
migration in a dose-dependent manner. Moreover, when
administered to CD nu/nu mice xenotransplanted with the
same DePTC cell line, CLM3 proved to inhibit both tumor
growth and weight, without showing any appreciable toxicity.27
Taken together, these results were extremely encouraging, as
they affirmed CLM3 as a sound and viable lead candidate,
deserving of further development. Accordingly, guided by
preliminary docking studies performed on the target kinases, we
embarked in rational optimization of CLM3, with the purpose
of obtaining novel and effective dual VEGFR2/RET inhibitors.
In this work we present the synthesis of new derivatives of
CLM3, whose benzylamino substituent at position 4 of the
heterocyclic core was suitably lengthened through the insertion
of arylamido (8a−g, Chart 1), arylureido (9a−g, Chart 1), and
arylsulfonyl (10a−g, Chart 1) groups. These were chosen to
plainly comply with pharmacophore requirements of the ATP
binding sites of both VEGFR2 and RET, highlighted by the
aforementioned docking studies. All of the synthesized
compounds were evaluated for their inhibitory properties
against the target tyrosine kinases and for their antiproliferative
activity toward both endothelial and human thyroid cancer cell
lines.
CHEMISTRY
expressing endogenous RET, and the medullary TT (IC50
=
■
26.93 7.60 μM), expressing the mutated protein, required
higher concentrations to be inhibited in their growth. Tested
on papillary dedifferentiated thyroid cancer cell line (DePTC),
The target inhibitors 8a−g, 9a−g, and 10a−g were synthesized
as illustrated in Scheme 1. Reaction of phenethylhydrazine, 1,29
with the commercially available 2-(ethoxymethylene)-
1227
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
malononitrile, 2, gave the pyrazole derivative, 3, whose
cyclization with boiling formic acid provided the 1-phenethyl-
1H-pyrazolo[3,4-d]pyrimidin-4(5H)-one, 4. Treatment of 4
with phosphoryl trichloride in DMF afforded the corresponding
4-chloro derivative, 5, which led to the intermediate 6 by
reaction with (3-nitrophenyl)methanamine, in the presence of
triethylamine. Catalytic hydrogenation of 6 carried out with
10% Pd on carbon at room temperature and atmospheric
pressure gave the key amino derivative 7, which was then
treated with the suitable reactants to give the target inhibitors.
In particular, reaction with the appropriate aroyl chloride in the
presence of triethylamine led to compounds 8a−g, treatment
with the suitable isocyanate provided the urea derivatives 9a−g,
and addition of substituted sulfonyl chlorides afforded the
sulfonamido inhibitors 10a−g.
Table 1. Tyrosine Kinases Inhibitory activity of Derivatives
8a−g, 9a−g, and 10a−g
a
IC50, μM
compd
X
R
VEGFR2
RET
RESULTS AND DISCUSSION
8a
CO
C6H5
8.57
3.48
5.37
3.29
2.66
2.03
6.45
2.16
3.40
3.17
1.09
1.09
3.27
6.74
4.96
4.89
2.89
1.03
3.87
7.21
2.23
52.5
22.4
4.26
9.64
17.1
24.1
28.1
22.6
7.04
9.07
4.80
■
8b
CO
C6H4-4-Br
Structure-Based Lead Optimization. We started our
optimization campaign testing CLM3 in vitro on the human
recombinant target proteins, VEGFR2 and RET, to confirm its
ability to compete with the ATP site of the tyrosine kinases.
When an IC50 value of 52.5 μM against VEGFR2 was obtained
and the ability of CLM3 to halve the kinase activity of RET was
verified (Table 1), we then clarified its mode of binding to both
of the target proteins, exploiting preliminary docking
calculations. In the case of VEGFR2, Glide software30
positioned CLM3 into the X-ray structure of the protein
(PDB code 4AG831) so that the N2 of the pyrazolopyrimidine
ring can establish an H-bond with the C919 backbone, the
phenylethanamine moiety is surrounded by V848, K868, A866,
L889, L1035, and V916 side chains, while the phenylethyl
group forms hydrophobic contact with L840 and F918 residues.
As highlighted in Figure 1, the short substitution pattern on
position 4 of the heterocyclic core prevents CLM3 from
reaching the VEGFR2 regulatory domain pocket (RDP),
represented by the cleft enclosed by the juxtamembrane
domain of the receptor refolding into the catalytic site.
Accordingly, interactions with key E885 and D1046 residues,
often hooked by known VEGFR2 inhibitors from the literature,
like sorafenib,31 are precluded, and this observation can be
instrumental to improve CLM3 binding toward VEGFR2.
At the same time, docking of CLM3 in the ATP domain of
RET kinase (PDB code 2IVV32) resulted in the binding mode
shown in Figure 2. Specifically, in the lowest-energy binding
pose, the N5 atom of the pyrazolopyrimidine ring establishes an
H-bond with the A807 backbone, CLM3 central core is stacked
between A756, V738, and L881, while the phenylethyl moiety
occupies the N-lobe, thus establishing hydrophobic interactions
with V804, L802, I788, L779, and K758 side chains. The
phenylethanamine moiety is pretty solvent exposed, and this
position could reasonably account for the lower activity of
CLM3 toward RET, deduced through functional evaluation on
8305C and TT cell lines, expressing the target receptor kinase,
and on the human recombinant protein.
8c
CO
C6H4-4-F
8d
CO
C6H4-4-NO2
C6H4-4-OCH3
C6H3-3,4-diOCH3
CH2C6H5
8e
CO
8f
CO
8g
CO
9a
CONH
CONH
CONH
CONH
CONH
CONH
CONH
SO2
C6H5
9b
C6H4-4-Br
9c
C6H4-4-F
b
9d
C6H4-4-NO2
C6H4-4-OCH3
C6H3-3,4-diOCH3
CH2C6H5
37
9e
9.26
15.3
10.3
7.90
12.85
4.83
9f
9g
10a
10b
10c
10d
10e
10f
10g
CLM3
C6H5
SO2
C6H4-4-Br
SO2
C6H4-4-F
b
SO2
C6H4-4-NO2
C6H4-4-OCH3
C6H3-3,4-diOCH3
CH2C6H5
57
SO2
57.2
b
SO2
8
SO2
58.6
b
48
a
IC50 values represent the concentration required to produce 50%
enzyme inhibition. Standard error of the mean (SEM) is ≤10%.
Percentage of kinase inhibition at 100 μM test compound.
b
Figure 1. Putative binding pose of CLM3 (gold sticks) in the active
site of VEGFR-2 shown as blue cartoon and cyan sticks. H-bonds are
represented as black dashed lines and part of the backbone is in
transparent ribbon, for clarity reason.
Thus, binding modes obtained for CLM3 in both VEGFR
and RET sites clearly suggested the structural limitations of
CLM3, but at same time, they pave the way for a CLM3
rational modification. Actually, on the bases of the docking
results obtained with VEGFR2, we decided to functionalize the
meta position of the phenylethanamine ring with a number of
substituents able to fill the RDP of the protein. Accordingly,
differently para-substituted phenyl rings bridged with amide,
urea, or sulfonamide moieties, already exploited for the
obtainment of VEGFR2 inhibitors from the literature,23,33,34
were inserted on the ring. At same time, to confer major
flexibility to this substituent, thus allowing an easier
accommodation inside the RDP, the methyl group of the
phenylethanamine function was removed. On the basis of the
suggested CLM3 binding mode in the RET active site, the
1228
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
unaltered this excellent inhibitory property, as the resulting
compounds 9b−f showed IC50 values ranging from 1.09 to 3.40
μM. The lengthened 9g (IC50 = 6.74 μM) was active as well. As
regards RET, the significant activity of the unsubstitued 9a
(IC50 = 7.04 μM) was gradually reduced by the presence of a
bromine atom, as in 9b (IC50 = 9.07 μM), a methoxy (9e, IC50
= 9.26 μM), and a dimethoxy group (IC50 = 15.3 μM). The
observed trend reached the peak with 9d, bearing a nitro group,
which reduced slightly the kinase activity of the target protein.
On the contrary, the lengthened derivative 9g proved to be
almost equipotent to the unsubstituted parent compound 9a.
Also in this subseries, the presence of an halogen atom turned
out to be the best substitution pattern to obtain a potent dual
inhibitor, as compound 9c, bearing a para fluorine atom
(VEGFR2, IC50 = 3.17 μM, RET, IC50 = 4.80 μM), displayed
the best inhibitory profile.
Figure 2. Putative binding pose of CLM3 (gold sticks) in the active
site of RET kinase shown as green cartoons. H-bonds are represented
as black dashed lines and part of the backbone is in transparent ribbon,
for clarity reason.
The choice to modify the lead, CLM3, exploiting a
benzenesulfonamide group, as in 10a−g, led to potent
VEGFR2 inhibitors. Moving from the unsubstituted 10a
(IC50 = 4.96 μM) the insertion of either a bromine atom, as
introduction of polar substituents on the phenylethanamine
group was expected to be beneficial.
Synthesis and Functional Evaluation of Novel
Pyrazolo[3,4-d]pyrimidines. Guided by the preliminary
docking studies, we started our structural optimization
synthesizing compounds 8a−g, characterized by the presence
of a benzamide residue in the meta position of the N4-benzyl
group of CLM3. As reported in Table 1, listing functional
activities expressed as IC50 values, all the novel compounds
showed a remarkable enhancement in inhibitory efficacy against
the target proteins when compared to the starting lead. In the
case of VEGFR2, compound 8a (IC50 = 8.57 μM) displayed a
6-fold increase in inhibitory potency with respect to CLM3
(IC50 = 52.5 μM). The insertion of electron-withdrawing
substituents in the para position of the distal phenyl ring, as in
8b (IC50 = 3.48 μM), 8c (IC50 = 5.37 μM), and 8d (IC50 = 3.29
μM), enhanced the observed activity, but an even better raise in
inhibitory potency was obtained through the insertion of
electron-donating groups. Indeed, the presence of a methoxy
function, as in 8e (IC50 = 2.66 μM), gave a 20-fold gain in
inhibitory activity with respect to CLM3, and compound 8f,
bearing a dimethoxy substituent pattern, turned out to be the
best VEGFR2 inhibitor of the whole subseries, showing an IC50
value of 2.03 μM. Finally, an increase in the distance between
the phenyl ring and the amide linker through a methylene
spacer was well tolerated, with 8g (IC50 = 6.45 μM) being
almost equipotent to the unsubstituted parent 8a. As regard to
RET, a significant inhibitory efficacy was observed as well.
Moving from 8a (IC50 = 22.4 μM), the insertion of electron-
donating groups like methoxy (8e, IC50 = 24.1 μM) and
dimethoxy (8f, IC50 = 28.1 μM) did not affect the activity, and
in 10b (IC50 = 4.89 μM), or methoxy groups, as in 10e (IC50
=
3.87 μM) and 10f (IC50 = 7.21 μM), did not modify
significantly the inhibitory activity, while the selective presence
of a nitro group (10d, IC50 = 1.03 μM), as well as the insertion
of a methylene spacer (10g, IC50 = 2.23 μM) exerted a positive
effect on the observed efficacy. On the contrary, with respect to
the tyrosine kinase RET, the benzenesulfonamide moiety gave
rise to contrasting results. Actually, for the unsubstituted 10a
(IC50 = 7.90 μM), the presence of either a bromine atom (10b,
IC50 = 12.8 μM) or a nitro function (10d, 57% inhibition at 100
μM), a methoxy (10e, IC50 = 57.2 μM), and a dimethoxy group
(10f, 8% inhibition at 100 μM) gave rise to a progressive loss of
activity. The same was also true for the lengthened derivative
10g, showing an IC50 value of 58.6 μM. On the contrary, the
insertion of a fluorine atom restored a remarkable inhibitory
efficacy. Indeed, once again, also in this subseries, the presence
of this kind of substituent led to the best dual inhibitory
pattern, and compound 10c, with IC50 values of 2.89 and 4.83
μM against VEGFR2 and RET, respectively, turned out to be
the most effective one of all the benzenesulfonamide
derivatives.
The best performing derivatives, 8b, 9c, and 10c, were also
assayed on the mutant human recombinant RETV804L
,
occurring in both spontaneous and familial MTC, which
proved to be insensitive to the already known kinase inhibitors,
including the commercially approved vandetanib. Significantly,
all the test compounds were demonstrated to block the kinase
activity of the mutant protein, showing IC50 values ranging
from 5.47 to 9.31 μM (Table 2).
Functional Evaluation on HUVEC, TT, and HNDF Cell
Lines. Representative examples of the three subseries 8a−d,
9a,b, and 10a,g were tested on HUVEC and TT cells to explore
the same was also true for the lengthened derivative 8g (IC50
22.6 μM). On the contrary, the presence of electron-
withdrawing groups as in 8b (IC50 = 4.26 μM), 8c (IC50
=
=
9.64 μM), and 8d (IC50 = 17.1 μM) gave a 1.5- to 5-fold
increase of efficacy. Significantly, the presence of a bromine
atom in the para position of the distal ring guaranteed the
highest inhibitory activity against the target protein, with 8b
being the best dual inhibitor of the whole subseries.
Analogous results were obtained through the synthesis of
derivatives 9a−g, where the N4-benzyl group of the lead,
CLM3, was functionalized with a phenylureido moiety.
Concerning VEGFR2, all the compounds displayed a
remarkable activity. Actually, with respect to the lead, the
unsubstituted 9a (IC50 = 2.16 μM) showed an almost 25-fold
gain in efficacy. The different substituents in the para position
of its distal ring, independent of their electronic nature, keep
Table 2. RETV804L Inhibitory Data of Derivatives 8b, 9c, and
10c
a
compd
RETV804L IC50, μM
8b
9.31
7.96
5.47
9c
10c
a
IC50 values represent the concentration required to produce 50%
enzyme inhibition. Standard error of the mean (SEM) is ≤10%.
1229
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
their ability to inhibit proliferation of these selected lines. After
a 72-h exposure, all the test derivatives proved to block cells
growth in a concentration dependent manner (Table 3).
compounds showed IC50 values of 22.0 2.0 μM (mean SD)
and 23.6
1.2 μM (mean
SD), respectively. These IC50
values were almost 4-fold higher than the ones observed in the
TT tumor cell line, thus confirming the preferential activity of
the compounds on the cell line with mutated RET.
Table 3. Inhibitory Activity of Derivatives 8a−d, 9a,b, and
10a,g against HUVEC and TT Cell Lines
Structure−Activity Relationships and Docking Stud-
ies. All the compounds reported in Table 1 were docked in the
VEGFR2 active site. Here, we have shown and analyzed in
detail the binding mode of compound 10c (Figure 3a) as the
main representative of the three new series of VEGFR2
inhibitors. Binding modes obtained for CLM3 (Figure 1) and
10c (Figure 3a) in VEGFR2 were highly superimposable. The
main differences reside in further interactions with E885 and
D1046, established by the sulfonamide group of 10c (Figure
3a), and with L889, I892, V899, L1019, H1026 (RDP),
engaged with the pendent phenyl ring, which accounts for the
11-fold increase in inhibitory potency observed for 10c when
compared to CLM3. Notably, all the novel derivatives,
regardless their polar bridge groups, CO, CONH, and SO2,
showed highly similar pattern of interactions as described for
10c (see Figure 1-SI in Supporting Information), thus
legitimating the gain in inhibitory potency of all the novel
derivatives with respect to CLM3. Specifically, in the 8a−g
subseries, the short bridge did not allow a complete fill of the
RDP. Indeed, compounds 8a was less active than 9a and 10a,
but the introduction of 4-substituents on the distant phenyl
ring further improved the inhibitory potency. Regarding the
RET kinase, binding modes obtained for the most active
compounds of the amide and urea series, 8b and 9c,
respectively (Figure 4), were pretty superimposable with that
of CLM3 (Figure 2), with main differences residing in further
possible interactions of amide and urea linkers and the terminal
phenyl ring lodged near the region of where the AMP
phosphate group binds (PDB code 2IVT,32 Figure 4a). Indeed,
both linkers are in proximity to S811 and an H-bond is likely to
occur. Meanwhile the 4-substituted benzene ring is just above
the N879 amidic side chain, thus justifying a general
improvement in the IC50 values toward RET. Furthermore,
the substitution of the aforementioned benzene with a bromine
atom, in the case of the shorter amide linker (8b), and with a
fluorine atom, in the case of the longer urea bridge (9c), gave
the possibility of establishing an H-bond with the NH group of
the E734, thus producing lower IC50 values with respect to their
nonhalogenated counterparts (Figure 4a and Figure 4b).
Finally, docking of 10c (Figure 3b) shown that the sulfonamide
moiety establishes multiple H-bonds with the hydroxyl and
ammonium groups of the Y806, K728, and K740 side chains,
respectively. However, the peculiar geometry of the sulfona-
mide group forces the phenyl ring to occupy a different portion,
a
IC50, μM
compd
HUVEC
TT
8a
15.31 0.59
9.83 0.97
13.82 1.32
13.48 2.30
13.57 3.88
6.13 1.74
8.42 1.26
9.86 0.72
8.08 1.10
5.97 1.17
8b
8c
15.93 1.31
16.64 0.88
13.43 4.55
18.55 0.007
22.64 1.13
50.73 9.25
8d
9a
9b
10a
10g
a
IC50 SD values represent the concentration of drug that decreased
cell count by 50%.
Significantly, when compared to CLM3, the novel compounds
determined a greater and very promising activity on TT cell
proliferation, showing IC50 values in the low micromolar range
and a 5-fold gain of efficacy with respect to the parent lead. The
functional efficacy observed on TT cell lines was further
investigated by evaluating the ability of the most active
derivatives, namely, 8d, 9a, and 10a,g, to inhibit the
phosphorylation of the target receptor. After exposure to the
test compounds, the ratio of phosphorylated/nonphosphory-
lated RET protein of treated cells was significantly decreased in
TT cell lines (Table 4), thus confirming that the antiprolifer-
ative activity of the compounds is mainly due to their inhibition
of phosphorylation of RET.
Table 4. Inhibitory Activity of RET Phosphorilation of
a
Derivatives 8d, 9a, 10a, and 10g
compd
% of (pRET)/(total RET) vs 100% of control
8d
66.4 4.9*
64.5 12.6*
88.7 4.4
9a
10a
10g
68.9 1.7*
a
Results were obtained from experiments using compounds at their
respective antiproliferative IC50 values: mean SD, (∗) P < 0.05 vs
control.
Finally, compounds 8d and 10g were also tested on cycling
normal human fibroblast HNDF line to investigate their
antiproliferative activity on a noncancerous cell line. Test
Figure 3. Binding pose of 10c (gold sticks) in the active site of VEGFR-2 (a) and RET (b) shown as blue and green cartoons, respectively. H-bonds
are shown as black dashed lines and part of the backbone is in transparent ribbon, for clarity reason.
1230
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
Figure 4. Binding pose of 8b (a) and 9c (b) represented as gold sticks in the active site of RET, shown as green cartoon. H-bonds are shown as black
dashed lines and part of the backbone is in transparent ribbon, for clarity reason. Cocrystallyzed AMP molecule is superimposed with 8b (a).
where it is partially solvent exposed, thus accounting for the
lower activity of 10e−10g, while a general improved activity is
observed for the halogenated compounds, as for 10c.
vandetanib which is devoid of any inhibitory efficacy against
this form of protein.
A preliminary evaluation on both endothelial and medullary
thyroid cancer cell lines revealed the promising antiproliferative
profile of the novel derivatives. Compared to the lead, CLM3,
the test compounds exhibited a remarkable increase in efficacy
against the target thyroid cancer cell line TT, expressing
mutated RET, showing an almost 5-fold decrease of IC50 values,
as in 8d (6.13 1.74 μM) and 10g (5.97 1.17 μM). Despite
a concurrent inflection in the antiproliferative activity against
the HUVEC cell lines, the novel compounds possess a better
balanced dual target profile, thus improving, in principle, their
therapeutic efficacy against the complex network of proteins
accountable for both the pathogenesis and progression of TCs.
Interestingly, the functional efficacy observed for the novel
compounds here described is fully comparable to the ones
shown by the already commercially approved tyrosine kinase
inhibitors cabozantinib and vandetanib, tested by Verbeek and
co-workers35 in analogous assays conditions.
Finally, with the aim to gain information about the capability
of 8b, 9c, and 10c to bind RETV804L, docking was performed
using the mutated RET as the target receptor, showing that the
flexible phenylethyl moiety is able to shift its position with
respect to that observed when calculations are performed on
RET wild type. Particularly, when the three compounds are
docked in RETV804L, the phenylethyl branch rearranges itself
and occupies the region where AMP phosphate group binds,
thus losing interactions with L779 and L802 side chains and, at
the same time, gaining hydrophobic contacts with V738 and
L881 side chains. Obviously, while for 10c (Figure 5a) the
Further studies in animal models will prove their robustness
as prototypical drug candidates exploitable for the treatment of
TCs, including the medullary thyroid cancer, which is caused by
RET activating mutations and still needs a viable and effective
therapeutic solution.
Figure 5. Superposition of the docking binding poses of 10c (a) and
9c (b) into the WT (golden sticks) and V804L mutant (magenta
sticks) RET kinase. The active site of RET is shown as green cartoon
although part of the backbone is in transparent ribbon, for clarity
reason. The L804 residue is represented as balls and sticks.
EXPERIMENTAL SECTION
■
Chemistry. Melting points were determined using a Reichert
Kofler hot-stage apparatus and are uncorrected. Routine 1H NMR
̈
spectra were recorded in DMSO-d6 solution on a Varian Gemini 200
spectrometer operating at 200 MHz. Mass spectra were obtained on a
Hewlett-Packard 5988 A spectrometer using a direct injection probe
and an electron beam energy of 70 eV. Evaporation was performed in
vacuo (rotary evaporator). Analytical TLC was carried out on Merck
0.2 mm precoated silica gel aluminum sheets (60 F-254). Microwave-
assisted reactions were carried out in sealed vessels using a Biotage
initiator 2.5 microwave synthesizer. Purity of the target inhibitors, 8a−
g, 9a−g, and 10a−g, was determined by HPLC analysis, using a Merck
Hitachi D-7000 liquid chromatograph (UV detection at 242 nm) and a
Discovery C18 column (250 mm × 4.6 mm, 5 μm, Supelco), with a
gradient of 30% water and 70% methanol and a flow rate of 1.5 mL/
min. All the compounds showed percent purity values of ≥95% (Table
7-SI, Supporting Information).
2-(Ethoxymethylene)malononitrile, 3-nitrobenzylamine hydrochlor-
ide, and the suitably substituted aroyl chlorides, arylsulfonyl chlorides,
and aryl isocyanates, used to obtain the target inhibitors, were from
Alfa Aesar, Aldrich, and Fluka.
Synthesis of 5-Amino-1-phenethyl-1H-pyrazole-4-carbon-
itrile, 3. Commercially available 2-(ethoxymethylene)malononitrile
(2, 1.22 g, 10.0 mmol) was added portionwise to an ice-cooled
solution of phenethylhydrazine (1, 1.36 g, 10.0 mmol) in ethanol.
Once the addition was complete, the reaction mixture was refluxed
under stirring until the disappearance of the starting materials (TLC
novel position of the phenylethyl moiety does not affect the
rest of the binding mode, in the case of 8b and 9c, the amide
and ureidic linkers with the distant phenyl ring have to readjust
altogether, and as shown in Figure 5b, these branches come to
occupy the same region of the sulfonamide moiety of 10c.
CONCLUDING REMARKS
■
In this work we presented a successful example of structure-
based optimization, pursued on the known tyrosine kinase
inhibitor CLM3.27,28 Guided by preliminary docking studies,
we rationally explored different substitution patterns on the
heterocyclic core, with the purpose of improving the dual
inhibitory profile of the starting lead.
The novel synthesized compounds displayed a remarkable
increase in inhibitory potency against the target proteins,
showing IC50 values in the micromolar range. Significantly, the
best performing ones, 8b, 9c, and 10c, proved to inhibit also
the mutant RETV804L which, to date, is the main cause of
functional failure of the well-known tyrosine kinase inhibitors
from the literature, including the commercially approved
1231
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
analysis). After the mixture was cooled to room temperature, the solid
was then poured into crushed ice, and the solid obtained was collected
by filtration and purified through crystallization from the suitable
solvent and characterized with physicochemical and spectroscopic data
(Supporting Information, Tables 3-SI and 4-SI).
was collected by filtration and purified by crystallization from toluene.
1
Mp 166−170 °C. Yield: 56%. H NMR (δ, ppm): 7.53 (s, 1H, H3),
7.27−7.18 (m, 5H, ArH), 6.56 (s, 2H, NH2, exc), 4.08 (t, 2H, J = 8.06
Hz), 2.93 (t, 2H, J = 8.06 Hz).
General Procedure for the Synthesis of N-(3-((1-Phenethyl-
1H-pyrazolo[3,4-d]pyrimidin-4-ylamino)methyl)phenyl)-
sulfonamide, 10a−g. A solution of N-(3-aminobenzyl)-1-phenethyl-
1H-pyrazolo[3,4-d]pyrimidin-4-amine 7 (0.344 g, 1.00 mmol) and the
suitable sulfonyl chloride (1.00 mmol) in DMF solution was used.
Then pyridine, 1 mL, was added and the mixture was left under
stirring at room temperature until the disappearance of the starting
materials (TLC analysis: 8/2 AcOEt/petroleum ether, 60−80 °C).
Once the reaction was complete, the crude was ice-cooled and 1 M
HCl was added. The solution was extracted with AcOEt, dried over
MgSO4, evaporated to dryness, and triturated with diethyl ether. The
resulting solid was collected by filtration and recrystallized from the
suitable solvent. (Supporting Information, Tables 5-SI and 6-SI).
Biology. Materials and Methods. Human recombinant protein
tyrosine kinases VEGFR2, RET and RETV804L, and Omnia Tyr peptide
7 kit were from Invitrogen. All solvents were from Sigma-Aldrich.
Tyrosine Kinase Assays. Assays were performed in 96-well
microtiter plates using the Omnia Tyr peptide 7 kit and following a
previously reported protocol.36,37 Tests were carried out at 30 °C in a
reaction mixture containing 5 μL of tyrosine kinase reaction buffer, 5
μL of tyrosine kinase substrate, 5 μL of 1 mM ATP, 5 μL of 1 mM
DTT, 25 μL of ultrapure water, and 5 μL of 3 mU/μL of the target
kinase, in a total volume of 50 μL. All the above reagents, except the
protein, were incubated at 30 °C for 5 min. Protein was then added to
start the reaction, which was monitored with the fluorescence meter
Victor3 PerkinElmer at 360 nm (excitation) and 485 nm (emission).
Kinase activity was calculated from a linear least-squares fit of the data
for fluorescence intensity versus time.
Enzymatic Inhibition. The inhibitory activity of the test
compounds was assayed by adding 5 μL of the inhibitor solution to
the reaction mixture described above. All the products were dissolved
in 100% DMSO and diluted to the appropriate concentrations with
tyrosine kinase reaction buffer, provided by the kit. Final concentration
of DMSO in assay solutions never exceed 1% and proved to have no
effects on protein activity. The inhibitory effect of the products was
routinely estimated at 100 μM. Those compounds found to be active
were then tested at additional concentrations between 100 μM and 10
nM. For a proper comparison, vandetanib was employed as the
reference standard. The determination of the IC50 values was
performed by linear regression analysis of the log dose response
curve, which was generated using at least five concentrations of the
inhibitor causing an inhibition between 20% and 80%, with two
replicates at each concentration. The 95% confidence limits (95% CL)
were calculated from t values for n − 2, where n is the total number of
determinations (Tables 1 and 2).
Cell Cultures and Viability Assays. Reagents and Cell Lines.
Test compounds were dissolved in a stock solution of 10 mM in 100%
dimethylsulfoxide for in vitro studies. The DMSO concentration in
control media was the one used to dilute the highest concentration of
test compounds in the same experiment. Recombinant human
epidermal growth factor (EGF) and basic fibroblast growth factor
(bFGF) were from PeproTechEC Ltd. (London, U.K.). Cell culture
media MCDB131 and F-12K, fetal bovine serum (FBS), L-glutamine,
penicillin, streptomycin, and antibiotics were from Gibco (Gaithers-
burg, MD). Type A gelatin from porcine skin, supplements, and all
other chemicals not listed in this section were obtained from Sigma
Chemical Co. (St. Louis, MO). Plastics for cell culture were supplied
by Sarsted (Verona, Italy). Human umbilical vein endothelial cells
(HUVEC; Clonetics, San Diego, CA) were maintained in MCDB131
culture medium supplemented with 10% heat-inactivated FBS, L-
glutamine 2 mM, heparin 10 units/mL, EGF 10 ng/mL, and bFGF 5
ng/mL and kept in a humidified atmosphere of 5% CO2 at 37 °C.
Human medullary thyroid cancer cell line (TT; ATCC, Manassas, VA,
USA) was maintained in 10% FBS F-12K medium supplemented with
L-glutamine 2 mM and kept in a humidified atmosphere of 5% CO2 at
37 °C. Normal human fibroblast cell line (HNDF) was maintained in
Synthesis of 1-Phenethyl-1H-pyrazolo[3,4-d]pyrimidin-
4(5H)-one, 4. 5-Amino-1-phenethyl-1H-pyrazole-4-carbonitrile, 4
(0.21 g, 1.00 mmol) was suspended in formic acid (0.5 mL) and
heated overnight under stirring at T = 120 °C. The resulting mixture
was poured into crushed ice, and the separated solid was collected by
filtration and purified by crystallization from methanol. Mp 244−248
(dec). Yield: 78%. 1H NMR (δ, ppm): 12.10 (s, 1H, NH, exc), 8.05 (s,
1H, H3), 7.99 (s, 1H, H6), 7.20−7.06 (m, 5H, ArH), 4.49 (t, 2H, J =
7.08 Hz), 3.12 (t, 2H, J = 7.08 Hz).
Synthesis of 4-Chloro-1-phenethyl-1H-pyrazolo[3,4-d]-
pyrimidine, 5. Vielsmeier complex (POCl3, 40.0 mmol, and DMF,
40.0 mmol) was added dropwise to an ice-cooled solution of 1-
phenethyl-1H-pyrazolo[3,4-d]pyrimidin-4(5H)-one (4, 10.0 mmol) in
anhydrous dichloromethane. Once addition was complete, the mixture
was refluxed under stirring until the disappearance of the starting
material (24 h, TLC analysis, 5/5 AcOEt/petroleum ether, 60−80
°C). The mixture was then ice-cooled, and the excess of POCl3 was
quenched with water. The crude was extracted with dichloromethane,
dried over MgSO4, and evaporated to dyness in vacuo to afford the
target 4-chloro-1-phenethyl-1H-pyrazolo[3,4-d]pyrimidine, 5, which
was exploited in the following reaction without further purification.
Synthesis of N-(3-Nitrobenzyl)-1-phenethyl-1H-pyrazolo-
[3,4-d]pyrimidin-4-amine, 6. 4-Chloro-1-phenethyl-1H-pyrazolo-
[3,4-d]pyrimidine 5 (0.258 g, 1.0 mmol), 3-nitrobenzylamine
hydrochloride (0.188 g, 1.0 mmol), and Et3N (0.278 mL, 2.0 mmol)
were suspended in toluene and irradiated with microwave at a
temperature of 80 °C for 5 min. The cooled residue was poured into
crushed ice and the solid separated was collected by filtration and
purified by crystallization from toluene to obtain the desired derivative,
6, as a yellow solid. Mp 113−114 °C. Yield: 61%. 1H NMR (δ, ppm):
8.90 (t, 1H, NH, J = 5.86 Hz, exc), 8.20−8.09 (m, 4H), 7.80 (d, 1H, J
= 7.30 Hz), 7.62 (t, 1H, J = 7.81 Hz), 7.24−7.10 (m, 5H), 4.84 (d, 2H,
J = 5.86 Hz), 4.51 (t, 2H, J = 7.08 Hz), 3.13 (t, 2H, J = 7.08 Hz).
Synthesis of N-(3-Aminobenzyl)-1-phenethyl-1H-pyrazolo-
[3,4-d]pyrimidin-4-amine, 7. A solution of N-(3-nitrobenzyl)-1-
phenethyl-1H-pyrazolo[3,4-d]pyrimidin-4-amine, 6 (0.374 g, 1.00
mmol), in absolute ethanol was added by 10% palladium on carbon
(0.10 mmol) and hydrogenated at atmospheric pressure and room
temperature until the disappearance of the starting material (TLC
analysis). After filtration of the catalyst, the solvent was evaporated to
dryness to give the target derivative, 7, as a white solid which was
purified by crystallization from toluene. Mp 120−122 °C. Yield: 83%.
1H NMR (δ, ppm): 8.63 (t, 1H, exc, J = 5.86 Hz), 8.18 (s, 1H), 8.13
(s, 1H), 7.20−7.14 (m, 5H), 6.93 (t, 1H, J = 7.32 Hz), 6.49−6.39 (m,
3H), 5.02 (bs, 2H, NH2, exc), 4.56 (d, 2H, J = 5.86 Hz), 4.49 (t, 2H, J
= 6.84 Hz), 3.13 (t, 2H, J = 6.10 Hz).
General Procedure for the Synthesis of N-(3-((1-Phenethyl-
1H-pyrazolo[3,4-d]pyrimidin-4-ylamino)methyl)phenyl)-4-sub-
stituted-benzamide, 8a−g. N-(3-Aminobenzyl)-1-phenethyl-1H-
pyrazolo[3,4-d]pyrimidin-4-amine 7 (0.344 g, 1.00 mmol) was
suspended either in anhydrous toluene (to obtain derivatives 8a−d)
or in anhydrous xylene (to obtain derivatives 8e−g), and to it were
added the suitable benzoyl chloride (1.00 mmol), Et3N (1.00 mmol),
and DMAP (12 mg, 0.1 mmol). The resulting mixture was refluxed
under stirring until the disappearance of the starting material (TLC
analysis). The crude so obtained was evaporated to dryness, and water
was added. The resulting compound, separated as a white solid, was
collected by filtration, purified through flash chromatography (eluting
system 5/5 AcOEt/petroleum ether, 60−80 °C), and crystallized from
the suitable solveny (Supporting Information, Tables 1-SI and 2-SI).
General Procedure for the Synthesis of 1-(3-((1-Phenethyl-
1H-pyrazolo[3,4-d]pyrimidin-4-ylamino)methyl)phenyl)-3-
urea, 9a−g. A suspension of N-(3-aminobenzyl)-1-phenethyl-1H-
pyrazolo[3,4-d]pyrimidin-4-amine 7 (0.344 g, 1.00 mmol) and the
suitable isocyanate (1.00 mmol) in toluene was irradiated with
microwave at a temperature of 90 °C for 5 min. The cooled residue
1232
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
10% FBS fibroblast basal medium supplemented with L-glutamine 2
mM and kept in a humidified atmosphere of 5% CO2 at 37 °C.
Proliferation Assay. In vitro chemosensitivity was tested on
HUVEC, TT, and HNDF cell lines. Cells were plated in sterile 24-well
plastic plates (which were coated with 1% of gelatin in the case of
endothelial cells) and treated for 72 h (using 3 × 104 cells/well of
normal and cancer cells, in 1 mL of medium) with added test
compounds (50, 10, 5, 1, and 0.1 μM) or with added vehicle (DMSO)
alone. At the end of the experiment, cells were harvested with trypsin/
EDTA, and viable cells were quantified using the automatic cell
counter ADAM MC digital B (Twin Helix, Milano, Italy). The data are
presented as the percentage of vehicle-treated cells. The concentration
of drugs that decreased cell count by 50% (IC50) compared with
controls was calculated by nonlinear fitting of experimental data. All
experiments were repeated, independently, three times with at least
nine samples for each concentration (Table 3).
orientation. Cross-docking experiments showed that 2IVV can allocate
also larger inhibitors like vandetenib, while 2X2L does not. Thus, 2IVV
was chosen for our docking experiments.
Dockings. The binding mode of CLM3 was studied by means of
docking experiments, with the Glide tool available in the Maestro
Package of Schrodinger, version 9.1.39 The 3D structure of the ligand
was generated with the Maestro fragment Build tool and then
geometrically optimized with Macromodel. The VEGFR-2 and RET
proteins structures were prepared through the Protein Preparation
Wizard of the Maestro 9.1 graphical user interface which assigns bond
orders, adds hydrogen atoms, deletes water molecules, and generates
appropriate protonation states.39
The docking grid box was centered on the reference ligand for each
X-ray structure, and docking runs were carried out using the standard
precision (SP) method. The selected docking pose was minimized
using OPLSA2005 as force field, the PRCG methods until a gradient
of 0.001 kcal/(mol·Å2) was obtained, and the implicit water model
implemented in Macromodel.40 Figures were rendered by the Chimera
software package.40
Cell-Based Phospho-RET Inhibition Assay. TT cells (5 × 104
cells/well) were seeded and maintained with 1% FBS medium. After
24 h, cells were treated continuously for 72 h with test compounds at a
concentration corresponding to the experimental IC50 of cell
proliferation. At the end of the experiment, the medium was removed
and cells were rinsed with ice-cold PBS and directly lysed with 0.5 mL
of ice-cold 1× lysis buffer (20 mM Tris, pH 7.5, 150 mM, 1 mM
EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium
pyrophosphate, 1 mM β-glycerol phosphate, 1 mM Na3VO4, 1 μg/
mL leupetin; Cell Signaling Technology, MA, USA, catalog no. 9803)
and 1 mM PMSF to each plate for 5 min at 4 °C. Lysates were
collected and sonicated on ice for 10 s. The samples were
microcentrifuged for 10 min at 4 °C, and the supernatant was
collected. Cell lysates were assayed per manufacturer’s instructions
with PathScan1 phospo-ret (panTyr) and Total Ret sandwich ELISA
kits (Cell Signaling Technology). The optical density was determined
using the microplate reader Multiskan Spectrum (Thermo Labsystems,
Milan, Italy) set to 450 nm. All experiments were repeated
independently six times with at least nine samples for each
concentration (Table 4).
ASSOCIATED CONTENT
* Supporting Information
■
S
Tables 1-SI to 9-SI, including physical, spectral, computational,
and purity data of compounds described, and Figure 1-SI,
including putative binding poses of 8b and 9c in the active site
of VEGFR-2. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Phone: (+)39 050 2219593. Fax: (+)39 050 2219605. E-mail:
concettina.lamotta@farm.unipi.it.
■
Notes
The authors declare no competing financial interest.
Molecular Modeling. VEGFR-2 and RET X-ray Selection. So
far, more than 30 X-ray structures are available in PDB for the
VEGFR2 kinase. A detailed comparison of all these structures shows
that, depending on the inhibitor type, the tyrosine kinase domain of
VEGFR2 undergoes important conformational rearrangements: (i) at
DFG activation loop level (DFGin = loop in active conformation,
inhibitor type I, and DFGout = loop in inactive conformation, inhibitor
type II); (ii) at the juxtamembrane level (JMin = autoinhibitory
conformation, inhibitor type IV; JMout = nonautoinhibitory con-
formation, inhibitor type II).33
Thus, we initially selected for docking studies eight X-ray structures
representative for the enzyme flexibility (Supporting Information,
Table 8-SI), preferring those cocrystallized with ligands similar to our
reference compound (CLM3, Chart 1). Particularly, we chose 1YWN
as representative of the DFGin conformation, while 2XIR, 3VHE,
4ASE, 4AGC, 4ASD, 4AGD, and 4AG8 have been chosen to represent
the DFGout and JMin/JMout conformations. Among these, the structure
4AG8 was used because it was demonstrated to be the best performing
one in Glide cross-docking experiments. In particular, employing this
structure, Glide was able to reproduce almost all the X-ray
conformations of all the analyzed cocrystal ligands with an average
root mean squared deviation (rmsd) of 3.33 Å (Supporting
Information, Table 9-SI).
ACKNOWLEDGMENTS
The study has been funded in part by the Ministero
dell’Istruzione, dell’Universita
C.L.M. and in part by the Associazione Italiana per la Ricerca
sul Cancro (AIRC) to G.B.
■
̀
e della Ricerca (MIUR), to
ABBREVIATIONS USED
■
TC, thyroid carcinoma; MTC, medullary thyroid carcinoma;
PTC, papillary thyroid carcinoma; DePTC, dedifferentiated
thyroid carcinoma; EGFR, epidermal growth factor receptor;
RET, REarranged during Transfection; VEGF, vascular
endothelial growth factor; VEGFR2, vascular endothelial
growth factor receptor 2; ATP, adenosine triphosphate; RDP,
regulatory domain pocket
REFERENCES
■
(1) Ito, Y.; Nikiforov, Y. E.; Schlumberger, M.; Vigneri, R. Increasing
incidence of thyroid cancer: controversies explored. Nat. Rev.
Endocrinol. 2013, 9, 178−184.
(2) Aschebrook-Kilfoy, B.; Schechter, R. B.; Shih, Y. C.; Kaplan, E. L.;
Chiu, B. C.; Angelos, P.; Grogan, R. H. The clinical and economic
burden of a sustained increase in thyroid cancer incidence. Cancer
Epidemiol. Biomarkers Prev. 2013, 22, 1252−1259.
As for the RET X-ray selection, a detailed comparison among the
phosphorylated structures currently available, superposing on the α
carbon atoms using 2IVV32 as the reference, indicated that the protein
folding was highly preserved. A further analysis of two structures with
the highest resolutions (2IVV and 2X2L38), both cocrystallized with an
organic inhibitor, reveals that the K758 can assume different
orientations upon ligand binding. In fact, in 2X2L, the phenyl-
methylidenedihydroindolone inhibitor does not occupy the N-lobe
hydrophobic cavity, and the K758 partially blocks the pocket entrance.
Conversely, the PP1 inhibitor in 2IVV enters the small cavity with its
methylphenyl moiety forcing the long K758 side chain to shift its
(3) Boufraqech, M.; Patel, D.; Xiong, Y.; Kebebew, E. Diagnosis of
thyroid cancer: state of art. Expert Opin. Med. Diagn. 2013, 7, 331−
342.
(4) Vini, L.; Harmer, C. Management of thyroid cancer. Lancet Oncol.
2002, 3, 407−414.
(5) Stassi, G.; Todaro, M.; Zerilli, M.; Ricci-Vitiani, L.; Di Liberto,
D.; Patti, M.; Florena, A.; Di Gaudio, F.; Di Gesu, G.; De Maria, M.
̀
1233
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
Thyroid cancer resistance to chemotherapeutic drugs via autocrine
production of interleukin-4 and interleukin-10. Cancer Res. 2003, 63,
6784−6790.
(6) Ye, L.; Santarpia, L.; Gagel, R. F. The evolving field of tyrosine
kinase inhibitors in the treatment of endocrine tumors. Endocr. Rev.
2010, 31, 578−599.
(7) Elisei, R.; Cosci, B.; Romei, C.; Bottici, V.; Ronzini, G.; Molinaro,
E.; Agate, L.; Vivaldi, A.; Faviana, P.; Basolo, F.; Miccoli, P.; Berti, P.;
Pacini, F.; Pinchera, A. Prognostic significance of somatic RET
oncogene mutations in sporadic medullary thyroid cancer: a 10-year
follow-up study. J Clin. Endocrinol. Metab. 2008, 93, 682−687.
(8) Lanzi, C.; Cassinelli, G.; Nicolini, V.; Zunino, F. Targeting RET
for thyroid cancer therapy. Biochem. Pharmacol. 2009, 77, 297−309.
(9) Biglietto, G.; Maglione, D.; Rambaldi, M.; Cerutti, J.; Romano,
A.; Trapasso, F.; Fedele, M.; Ippolito, P.; Chiappetta, G.; Botti, G.
Upregulation of vascular endothelial growth factor (VEGF) and
downregulation of placenta growth factor (PlGF) associated with
malignancy in human thyroid tumors and cell lines. Oncogene 1995, 11,
1569−1579.
(10) Knauf, J. A. Does the epidermal growth factor receptor play a
role in the progression of thyroid cancer? Thyroid. 2011, 21, 1171−
1174.
(11) Ye, L.; Santarpia, L.; Gagel, R. F. Targeted therapy for endocrine
cancer: the medullary thyroid carcinoma paradigm. Endocr. Pract.
2009, 15, 597−604.
(12) Gild, M. L.; Bullock, M.; Robinson, B. G.; Clifton-Bligh, R.
Multikinase inhibitors: a new option for the treatment of thyroid
cancer. Nat. Rev. Endocrinol. 2011, 7, 617−624.
(13) Wilhelm, S. M.; Adnane, L.; Newell, P.; Villanueva, A.; Llovet, J.
M.; Lynch, M. Preclinical overview of sorafenib, a multikinase inhibitor
that targets both Raf and VEGF and PDGF receptor tyrosine kinase
signaling. Mol. Cancer Ther. 2008, 7, 3129−3140.
(14) Gan, H. K.; Seruga, B.; Knox, J. J. Sunitinib in solid tumors.
Expert Opin. Invest. Drugs 2009, 18, 821−834.
(15) Mologni, L.; Redaelli, S.; Morandi, A.; Plaza-Menacho, I.;
Gambacorti-Passerini, C. Ponatinib is a potent inhibitor of wild-type
and drug-resistant gatekeeper mutant RET kinase. Mol. Cell.
Endocrinol. 2013, 377, 1−6.
(16) Santos, E. S.; Gomez, J. E.; Raez, L. E. Invest. New Drugs 2012,
30, 1261−1269.
(17) Kelly, R. J.; Rixe, O. Axitiniba selective inhibitor of the
vascular endothelial growth factor (VEGF) receptor. Target Oncol.
2009, 4, 297−305.
(18) Bible, K. C.; Suman, V. J.; Menefee, M. E.; Smallridge, R. C.;
Molina, J. R.; Maples, W. J.; Karlin, N. J.; Traynor, A. M.; Kumar, P.;
Goh, B. C.; Lim, W. T.; Bossou, A. R.; Isham, C. R.; Webster, K. P.;
Kukla, A. K.; Bieber, C.; Burton, J. K.; Harris, P.; Erlichman, C.; Mayo
Phase 2 Consortium; Mayo Clinic Endocrine Malignances Disease
Oriented Group.. A multiinstitutional phase 2 trial of pazopanib
monotherapy in advanced anaplastic thyroid cancer. J. Clin. Endocrinol.
Metab. 2012, 97, 3179−3184.
(19) Durante, C.; Paciaroni, A.; Plasmati, K.; Trulli, F.; Filetti, S.
Vandetanib: opening a new treatment practice in advanced medullary
thyroid carcinoma. Endocrine 2013, 44, 334−342.
(20) Nagilla, M.; Brown, R. L.; Cohen, E. E. Cabozantinib for the
treatment of advanced medullary thyroid cancer. Adv Ther. 2012, 29,
925−934.
(21) Chaud, N. G.; Haddad, R. I. Vandetanib for the treatment of
medullary thyroid cancer. Clin. Cancer Res. 2012, 19, 1−6.
(22) Lombardo, F.; Baudin, E.; Chiefari, E.; Arturi, F.; Bardet, S.;
Caillou, B.; Conte, C.; Dallapiccola, B.; Giuffrida, D.; Bidart, J. M.;
Schlumberger, M.; Filetti, S. Familial medullary thyroid carcinoma:
clinical variability and low aggressiveness associated with RET
mutation at codon 804. J. Clin. Endocrinol. Metab. 2002, 87, 1674−
1680.
(23) Radi, M.; Tintori, C.; Musumeci, F.; Brullo, C.; Zamperini, C.;
Drenassi, E.; Fallacara, A.; Vignaioli, G.; Crespan, E.; Zanoli, S.;
Laurenzana, I.; Filippi, I.; Maga, G.; Schenone, S.; Angelucci, A.; Botta,
M. Design, synthesis, and biological evaluation of pyrazolo[3,4-
d]pyrimidines active in vivo on the Bcr-Abl T315I mutant. J. Med.
Chem. 2013, 56, 5382−5394.
(24) Yang, L. L.; Li, G. B.; Ma, S.; Zou, C.; Zhou, S.; Sun, Q. Z.;
Cheng, C.; Chen, X.; Wang, L. J.; Feng, S.; Li, L. L.; Yang, S. Y.
Structure−activity relationship studies of pyrazolo[3,4-d]pyrimidine
derivatives leading to the discovery of a novel multikinase inhibitor
that potently inhibits FLT3 and VEGFR2 and evaluation of its activity
against acute myeloid leukemia in vitro and in vivo. J. Med. Chem.
2013, 56, 1641−1655.
́
(25) Diner, P.; Alao, J. P.; Soderlund, J.; Sunnerhagen, P.; Grøtli, M.
̈
Preparation of 3-substituted-1-isopropyl-1H-pyrazolo[3,4-d]pyrimidin-
4-amines as RET kinase inhibitors. J. Med. Chem. 2012, 55, 4872−
4876.
(26) Le Brazidec, J. Y.; Pasis, A.; Tam, B.; Boykin, C.; Black, C.;
Wang, D.; Claassen, G.; Chong, J. H.; Chao, J.; Fan, J.; Nguyen, K.;
Silvian, L.; Ling, L.; Zhang, L.; Choi, M.; Teng, M.; Pathan, N.; Zhao,
S.; Li, T.; Taveras, A. Synthesis, SAR and biological evaluation of 1,6-
disubstituted-1H-pyrazolo[3,4-d]pyrimidines as dual inhibitors of
Aurora kinases and CDK1. Bioorg. Med. Chem. Lett. 2012, 22,
2070−2074.
(27) Antonelli, A.; Bocci, G.; La Motta, C.; Ferrari, S. M.; Fallahi, P.;
Fioravanti, A.; Sartini, S.; Minuto, M.; Piaggi, S.; Corti, A.; Alì, G.;
Berti, P.; Fontanini, G.; Danesi, R.; Da Settimo, F.; Miccoli, P. Novel
pyrazolopyrimidine derivatives as tyrosine kinase inhibitors with
antitumoral activity in vitro and in vivo in papillary dedifferentiated
thyroid cancer. J. Clin. Endocrinol. Metab. 2011, 96, E288−E296.
(28) Bocci, G.; Fioravanti, A.; La Motta, C.; Orlandi, P.; Canu, B.; Di
Desidero, T.; Mugnaini, L.; Sartini, S.; Cosconati, S.; Frati, R.;
Antonelli, A.; Berti, P.; Miccoli, P.; Da Settimo, F.; Danesi, R.
Antiproliferative and proapoptotic activity of CLM3, a novel multiple
tyrosine kinase inhibitor, alone and in combination with SN-38 on
endothelial and cancer cells. Biochem. Pharmacol. 2011, 81, 1309−
1316.
(29) Zhang, J.; Yin, Z.; Leonard, P.; Wu, J.; Sioson, K.; Liu, C.; Lapo,
R.; Zheng, S. A variation of the Fischer indolization involving
condensation of quinone monoketals and aliphatic hydrazines. Angew.
Chem., Int. Ed. 2013, 52, 1753−1757.
(30) Glide, version 5.5; Schrodinger, LLC: New York, 2009.
(31) McTigue, M.; Murray, B. W.; Chen, J. H.; Deng, Y. L.; Solowiej,
J.; Kania, R. S. Molecular conformations, interactions, and properties
associated with drug efficiency and clinical performance among
VEGFR TK inhibitors. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 18281−
18289.
(32) Knowles, P. P.; Murray-Rust, J.; Kjaer, S.; Scott, R. P.;
́
Hanrahan, S.; Santoro, M.; Ibanez, C. F.; McDonald, N. Q. Structure
̃
and chemical inhibition of the RET tyrosine kinase domain. J. Biol.
Chem. 2006, 281, 33577−33587.
(33) McTigue, M.; Murray, B. W.; Chen, J. H.; Deng, Y.-L.; Solowiej,
J.; Kania, R. S. Molecular conformations, interactions, and properties
associated with drug efficiency and clinical performance among
VEGFR TK inhibitors. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 18281−
18289.
(34) Kubo, K.; Shimizu, T.; Ohyama, S.; Murooka, H.; Iwai, A.;
Nakamura, K.; Hasegawa, K.; Kobayashi, Y.; Takahashi, N.; Takahashi,
K.; Kato, S.; Izawa, T.; Isoe, T. Novel potent orally active selective
VEGFR-2 tyrosine kinase inhibitors: synthesis, structure−activity
relationships, and antitumor activities of N-phenyl-N′-{4-(4-
quinolyloxy)phenyl}ureas. J. Med. Chem. 2005, 48, 1359−1366.
(35) Verbeek, H. H. G.; Alves, M. M.; de Groot, J.-V. B.; Osinga, J.;
Plukker, J. T. M.; Links, T. P.; Hofstra, R. M. W. The effects of four
different tyrosine kinase inhibitors on medullary and papillary thyroid
cancer cells. J. Clin. Endocrinol. Metab. 2011, 96, E991−E995.
(36) Perspicace, E.; Jouan-Hureaux, V.; Ragno, R.; Ballante, F.;
Sartini, S.; La Motta, C.; Da Settimo, F.; Chen, B.; Kirsch, G.;
Schneider, S.; Faivre, B.; Hesse, S. Design, synthesis and biological
evaluation of new classes of thieno[3,2-d]pyrimidinone and thieno-
[1,2,3]triazine as inhibitor of vascular endothelial growth factor
receptor-2 (VEGFR-2). Eur. J. Med. Chem. 2013, 63, 765−781.
1234
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235

Journal of Medicinal Chemistry
Article
(37) La Motta, C.; Sartini, S.; Tuccinardi, T.; Nerini, E.; Da Settimo,
F.; Martinelli, A. Computational studies of epidermal growth factor
receptor: docking reliability, three-dimensional quantitative structure−
activity relationship analysis, and virtual screening studies. J. Med.
Chem. 2007, 52, 964−975.
(38) Mologni, L.; Rostagno, R.; Brussolo, S.; Knowles, P. P.; Kjaer,
S.; Murray-Rust, J.; Rosso, E.; Zambon, A.; Scapozza, L.; Mcdonald, N.
Q.; Lucchini, V.; Gambacorti-Passerini, C. Synthesis, structure−activity
relationship and crystallographic studies of 3-substituted indolin-2-one
RET inhibitors. Bioorg. Med. Chem. 2010, 18, 1482−1496.
(39) Maestro, version 9.1; Schrodinger, LLC: New York, 2009;
̈
http://www.schrodinger.com/ (accessed May 15, 2011).
(40) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.;
Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. UCSF chimera: a
visualization system for exploratory research and analysis. J. Comput.
Chem. 2004, 25, 1605−1612.
1235
dx.doi.org/10.1021/jm401358b | J. Med. Chem. 2014, 57, 1225−1235