Discovery of the selective and efficacious inhibitors of FLT3 mutations

Article abstract of DOI:10.1016/j.ejmech.2018.06.010

Fms-like tyrosine kinase 3 (FLT3) is among the most frequently mutated protein in acute myeloid leukemia (AML), which has been confirmed as an important drug target for AML chemotherapy. Starting from the lead compound LT-106-175, a series of 1-H-pyrazole-3-carboxamide derivatives were synthesized to improve the FLT3 inhibitory potency and selectivity. Among them, compound 50 was identified as a highly potent and selective FLT3 inhibitor (IC₅₀ = 0.213 nM), which showed equal activities against various mutants of FLT3 including FLT3 (ITD)-D835V and FLT3 (ITD)-F691L that is resistant to quizartinib. Compound 50 also exhibited efficacy against the human AML cell line MV4-11 (IC₅₀ = 16.1 nM) harboring FLT3-ITD mutants. Inversely, compound 50 displayed no cytotoxicity to FLT3-independent cells, and the biochemical analyses showed that its effects were related to the inhibition of FLT3 signal pathways. Additionally, compound 50 induced apoptosis in MV4-11 cell as demonstrated by flow cytometry. Moreover, compound 50 showed enhanced metabolic stability. Altogether, it was concluded that compound 50 could be a promising FLT3 inhibitor for further developing therapeutic remedy of AML.

Full text of DOI:10.1016/j.ejmech.2018.06.010

Accepted Manuscript
Discovery of the selective and efficacious inhibitors of FLT3 mutations
Yanle Zhi, Baoquan Li, Chao Yao, Hongmei Li, Puzhou Chen, Jiyin Bao, Tianren Qin,
Yue Wang, Tao Lu, Shuai Lu
PII:
S0223-5234(18)30501-4
10.1016/j.ejmech.2018.06.010
EJMECH 10480
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 6 February 2018
Revised Date: 4 May 2018
Accepted Date: 4 June 2018
Please cite this article as: Y. Zhi, B. Li, C. Yao, H. Li, P. Chen, J. Bao, T. Qin, Y. Wang, T. Lu, S. Lu,
Discovery of the selective and efficacious inhibitors of FLT3 mutations, European Journal of Medicinal
Chemistry (2018), doi: 10.1016/j.ejmech.2018.06.010.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Discovery of the selective and efficacious inhibitors of FLT3 mutations
Yanle Zhia,b, Baoquan Lia, Chao Yaoa, Hongmei Lia, Puzhou Chena, Jiyin Baoa,
Tianren Qinc, Yue Wanga, Tao Lu*,a,b, Shuai Lu*,a,b
aSchool of Sciences, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing
210009, PR China.
bState Key Laboratory of Natural Medicines, China Pharmaceutical University, 24
Tongjiaxiang, Nanjing 210009, PR China.
cSchool of Pharmacy, China Pharmaceutical University, 639 Longmian Avenue,
Nanjing, 211198, PR China.
ABSTRACT
Fms-like tyrosine kinase 3 (FLT3) is among the most frequently mutated protein
in acute myeloid leukemia (AML), which has been confirmed as an important drug
target for AML chemotherapy. Starting from the lead compound LT-106-175, a series
of 1-H-pyrazole-3-carboxamide derivatives were synthesized to improve the FLT3
inhibitory potency and selectivity. Among them, compound 50 was identified as a
highly potent and selective FLT3 inhibitor (IC50 = 0.213 nM), which showed equal
activities against various mutants of FLT3 including FLT3 (ITD)-D835V and FLT3
(ITD)-F691L that is resistant to quizartinib. Compound 50 also exhibited efficacy
against the human AML cell line MV4-11 (IC50 = 16.1 nM) harboring FLT3-ITD
mutants. Inversely, compound 50 displayed no cytotoxicity to FLT3-independent cells,
and the biochemical analyses showed that its effects were related to the inhibition of
FLT3 signal pathways. Additionally, compound 50 induced apoptosis in MV4-11 cell

ACCEPTED MANUSCRIPT
as demonstrated by flow cytometry. Moreover, compound 50 showed enhanced
metabolic stability. Altogether, it was concluded that compound 50 could be a
promising FLT3 inhibitor for further developing therapeutic remedy of AML.
INTRODUCTION
Acute myeloid leukemia (AML) is characterized by the accumulation of a large
number of abnormal cells that fail to differentiate into the functional granulocytes or
monocytes[1]. Previous studies have found that elderly AML patients, which
constitutes the majority of AML population, generally had poor outcomes, high
healthcare expenditures, and median survival of <3 months[2]. Even with current
treatments, as much as 70% of patients of 65 years old or older died of AML within 1
year after diagnosis[2]. Hence, effective treatment of AML in elderly patients still
represents a severe challenge.
Fms-like tyrosine kinase 3 (FLT3), a member of the class III receptor tyrosine
kinase family, is expressed by the immature hematopoietic cells, playing an important
role in the proliferation, survival, and differentiation of multipotent stem cells[3].
However, FLT3 is over-expressed in 70% to 100% of AML cases[4-6] and mutation
of FLT3 is the most common molecular alteration in AML[7]. Internal tandem
duplication alteration (FLT3-ITD) in juxtamembrane portion is the most prevalent
FLT3 mutation occurring in approximately 25% of those with de novo AML[7]. Point
mutations in the kinase domain of the tyrosine (TKD) present in about 7% of AML
cases[8, 9]. Both ITD and TKD are activating mutations, causing ligand-independent
or constitutive activation of FLT3 signaling, and thereby promoting AML cell

ACCEPTED MANUSCRIPT
survival and proliferation, which are often associated with poor prognosis[7], higher
rate of relapse and reduced overall survival[8, 10-12]. Therefore, FLT3 has become
an important target for AML[13, 14].
So far, a number of FLT3 inhibitors have been developed[15]. First-generation
inhibitors, including sunitinib[16], sorafenib, midostaurin, lestaurtinib[17], and
tandutinib[18], lack the specificity for FLT3. Consequently, their drug dosage and
frequency of application may be limited by toxicity caused by the off-target activities;
in turn, compromising the effectiveness of the treatment[19, 20]. Almost all of these
compounds discontinued at phase II clinical trials, except midostaurin which was first
approved by FDA for the treatment of AML[21]. Comparatively, the
second-generation FLT3 inhibitors are more specific and potent, such as
quizartinib[22] (phase III), which is still under active development. However, its
c-Kit-inhibitory effect caused strong suppression of normal hematopoietic
stem/progenitor cells[23]. Furthermore, some TKD mutations of FLT3 (D835Y,
D835V, D835F, or F691L) could induce the resistant to quizartinib[24]. Thus,
small-molecule inhibitors of FLT3 with strong potency against both FLT3-ITD and
FLT3-TKD simultaneously and with high selectivity towards them hold great promise
for more effective chemotherapy of AML. Here, we report the development of a new
kind of FLT3 selective inhibitors against both FLT3-ITD and FLT3-TKD.

ACCEPTED MANUSCRIPT
Chemistry
Scheme 1. Synthetic routes of compounds 33-41.
.
Reagents and conditions: (a) R1-H, Et3N, CH2Cl2; (b) FeO(OH)/C, 80% NH2NH2 H2O, 95%
EtOH; (c) 4-Nitropyrazole-3-carboxylic acid or 1-methyl-4-nitro-1H-pyrazole-3-carboxylic acid
or 5-methyl-4-nitro-1H-pyrazole-3-carboxylic acid, EDC.HCl, HOBt, DMF, r.t.; (d) FeO(OH)/C,
.
80% NH2NH2 H2O, 95% EtOH; (e) R2-Cl, AcOH/H2O.
Compounds 33-41 were prepared by the general approach shown in Scheme 1.
4-Nitrobenzyl bromide was used as the starting material to synthesize compounds 1
and 2 through a nucleophilic substitution reaction. Reduction of compound 1 and 2
with 80% hydrazine hydrate gave compounds 3 and 4, which were then coupled with
4-nitropyrazole-3-carboxylic acid, 1-methyl-4-nitro-1H-pyrazole-3-carboxylic acid
and 5-methyl-4-nitro-1H-pyrazole-3-carboxylic acid, respectively[25, 26][25, 26], to
obtain compounds 5-8. The same reduction method was used to convert compounds
5-8 to the corresponding amines 9-12. The target compounds 33-41 were prepared by
the direct ammonolysis of certain heterochorides with compounds 9-12. Compounds
42-51 were prepared in a similar way (Scheme 2). Intermediates 13-15 were

ACCEPTED MANUSCRIPT
synthesized by the nucleophilic substitution of 1-fluoro-4-nitrobenzene with the
relative amines, and then the same procedures in Scheme 1 were followed to give
compounds 25-43. Final compounds 44-51 were obtained upon the deprotection by
acidic cleavage.
Scheme 2. Synthetic routes of compounds 42-51.
.
Reagents and conditions: (a) R1-H, K2CO3, DMSO; (b) FeO(OH)/C, 80% NH2NH2 H2O, 95%
EtOH; (c) 4-Nitropyrazole-3-carboxylic acid or 1-methyl-4-nitro-1H-pyrazole-3-carboxylic acid
or 5-methyl-4-nitro-1H-pyrazole-3-carboxylic acid, EDC.HCl, HOBt, DMF, r.t.; (d) FeO(OH)/C,
.
80% NH2NH2 H2O, 95% EtOH; (e) R2-Cl, AcOH/H2O, 50 °C; (f) CF3COOH, CH2Cl2.

ACCEPTED MANUSCRIPT
RESULTS AND DISCUSSION
In our previous studies, a multiple kinases inhibitor LT-106-175 was discovered,
which showed potent inhibitory activities against CDKs and FLT3. From the docking
models of LT-106-175 with CDK2/6 and FLT3, we found that LT-106-175 interacted
with CDK2/6 in the same manner. In the hinge area, the 1H-pyrazole-3-formyl motif
formed three hydrogen bonds with the protein, and the fused thieno[2,3-d]pyrimidine
ring was involved in the edge to face interaction with the gatekeeper residue Phe80/98,
which played an important role in CDK2/6 inhibition[27, 28]. In the FLT3 model,
three hydrogen bonds also presented between LT-106-175 and the hinge area;
however the fused thieno[2,3-d]pyrimidine ring was far from reside Phe80, and the
back pocket was bigger than that of CDK2/6. Thus, the back pocket of FLT3 could
accommodate bulky groups. Therefore, we assumed that the introduction of some
suitable groups to impede the edge to face interaction in CDK2/6 or to occupy more
space in the back pocket of FLT3 would be beneficial for FLT3 selectivity. We hoped
that the selectivity of this kind of compounds towards FLT3 could be enhanced
through the sophisticated optimization guided by the assumption mentioned above, so
that the potential side effects caused by CDK2 inhibition could be avoided[29].

ACCEPTED MANUSCRIPT
Figure 1. Molecular docking analysis of LT-106-175 bound to CDK2, CDK6 and FLT3 kinases.
(A) Compound LT-106-175 docked into CDK2 (PDB code 2VU3). (B) Compound LT-106-175
docked into CDK6 (PDB code 2EUF). (C) Compound LT-106-175 docked into homology model
of FLT3[30]. (D) Structure of compound LT-106-175.
As shown in Table 1, Table 2 and Table 3, 19 new compounds were presented.
All compounds were evaluated for their abilities to inhibit CDK2/6 and FLT3 at 0.123
µM firstly. Compounds that decreased the kinase activity by more than 80% were
selected to further determine their half-maximal inhibitory concentration (IC50)
against CDK2/6 or FLT3. Additionally, we performed the growth-inhibition assays
using human AML cell line MV4-11.
Methyl substitution onto the NH of pyrazole ring led to the total loss of activity
against CDKs and FLT3 for compound 33, which confirmed the importance of
hydrogen bonds in the hinge region. Similarly, the introduction of a methyl group to

ACCEPTED MANUSCRIPT
the C5-position of pyrazole ring also eliminated the inhibitory activity against both
CDK2/6 and FLT3 for compound 34. However, compounds 36 and 37 showed great
FLT3 selectivity over CDK2/6 when a methyl group was introduced to the pyrimidine
ring, which was consistent with the assumption deduced from the molecule docking
study. The IC50 value of compound 36 against FLT3 was 0.65 nM, which achieved
almost 100-fold selectivity towards FLT3 over CDK2/6, and the anti-proliferative
activity of compounds 36 and 37 against MV4-11 cell remained potent (IC50 = 0.052
nM and 0.062 nM, respectively), though the IC50 values decreased slightly when
compared with LT-106-175. This indicated that high FLT3 selectivity of our
compounds would not attenuate their cell activity significantly. Replacing the methyl
in 36 with cyclopropyl (38) could improve the FLT3 selectivity over CDK2/6, but it
also led to the 10-fold loss of potency against FLT3. Those data suggested that the
methyl substitution at position 2 of pyrimidine was currently optimal. In order to
investigate whether additional space exists for the introduction of groups, the
isopropyl groups were adopted at C2 position of thiophene (compounds 40 and 41).
Either the potency of FLT3 or the kinase selectivity was only maintained at the same
level. (Table 1), but this result suggested that the modification on the thiophene ring
was tolerated.

ACCEPTED MANUSCRIPT
Table 1. Structures and biological data of compounds 33-41
(% enzyme activity) a
(% cell growth
inhibition) c /IC50 (µM) d
MV4-11
Compou
/IC50 (nM) b
CDK6
Structure
nds
CDK2
FLT3
LT-106-
175
9.10 ± 0.16 3.33 ± 0.25 0.60 ± 0.11
0.016± 0.006
98.55 %
99.90 %
99.65 %
99.34 %
95.65 %
96.19 %
87.56%
98.52 %
98.52 %
nde
nd
nd
33
34
35
51.10 %
51.96 %
72.09 %
79.74 %
102.73 %
95.40 %
0.65 ± 0.02
3.63 ± 0.11
0.052 ± 0.004
0.062 ± 0.007
0.061 ± 0.007
0.078 ± 0.012
0.048 ± 0.006
36
37
38
39
40
15.7 ±
0.33
89.75 %
81.41 %
16.50 ± 0.45
0.66 ± 0.13
33.21 ±
1.28
25.4 ± 0.57

ACCEPTED MANUSCRIPT
19.24 ±
21.5 ± 0.64
0.73 ± 0.27
0.065 ± 0.002
41
0.93
a % Enzyme activity at 0.123 µM in the presence of 10 µM ATP; b In the presence of 10 µM ATP,
the values are the mean ± SD from three independent experiments. c % cell growth inhibition at 1
µΜ; d the highest test concentration is 1 µM, the values are the mean ± SD from three independent
experiments. e nd: not determined
Varying the substitution in hydrophilic area helped us to identify which group
was superior for the potency and selectivity of FLT3. Replacement of the
4-methylpiperazin in 36, 38, or 41 with morpholinyl (37, 39 and 40) was tolerated in
terms of FLT3 potency and selectivity. Consistently, those compounds also kept
moderate anti-proliferation activities in MV4-11 cell line. To our surprise, the
introduction of the methylpiperazin-1-yl (42) and morpholinyl (43) directly onto the
benzene ring could further improve FLT3 selectivity, indicating that the hydrophilic
group also contributed to the FLT3 selectivity. Moreover, the potency of these two
compounds against MV4-11 was also elevated mildly. Continuing modification on the
4-methylpiperazin gave compound 44, from which methyl substitution was removed.
Although this change was minor, compound 44 showed almost 131 times selectivity
for FLT3 over CDK2/6, and its cell growth inhibitory activity was equal to
LT-106-175. Overall, piperazin ring was the optimal hydrophilic group to achieve
high FLT3 selectivity and inhibitory potency.
On the basis of the optimized hydrophilic group, SAR of various substitutions at
pyrimidine ring was explored again. The activities against CDK2/6 decreased along

ACCEPTED MANUSCRIPT
with the bulk increase of the substituted groups (Table 2), such as ethyl (47),
isopropyl (48), tert-butyl (49), cyclopropyl (46) groups, etc. However, an obvious
decrease in FLT3 kinase inhibition also occurred for those compounds (IC50:
2.53-3.42 nM), accompanied by the reduction of cell activity, especially for
compounds 49 bearing larger substituted group (IC50: 20.5 nM). Only compound 45
with chlorine substitution exhibited maintained activities against FLT3 and MV4-11,
which showed that the bulk of group was vital. These results proved that methyl
group was the optimal option to keep the balance between FLT3 selectivity and cell
potency.
Table 2. Optimization of the hydrophilic group and the gatekeeper group.
( % cell growth
(% enzyme activity) a
inhibition) c
Comp
/IC50(nM) b
R1
R2
/IC50 (µM) d
MV4-11
ounds
CDK2
71.23 %
78.81 %
81.54 ±
1.23
CDK6
72.74 %
83.08 %
84.31 ±
1.55
FLT3
-CH3
-CH3
0.95 ± 0.12
3.80 ± 0.18
0.027 ± 0.003
0.033 ± 0.009
42
43
-CH3
-Cl
0.62 ± 0.04
0.47 ± 0.09
0.019 ± 0.002
0.026 ± 0.006
44
45
56.82 %
63.57 %

ACCEPTED MANUSCRIPT
97.44 %
86.59%
88%
2.53 ± 0.02
3.42 ± 0.12
11.31 ± 0.21
0.059 ± 0.005
0.065 ± 0.002
0.057 ± 0.004
0.098 ± 0.001
46
47
48
49
87.04%
93.50%
99.21%
100.89%
100.32%
20.51 ± 0.42
a
b
% Enzyme activity at 0.123 µM in the presence of 10 µM ATP; In the presence of 10 µM
ATP with 0.123 µM compound, the values are the mean ± SD from three independent experiments.
c % cell growth inhibition at 1 µΜ; d the highest test contraction is 1 µM, the values are the mean ±
SD from three independent experiments. e nd: not determined
As reported, compounds with pyrimidothiophene ring are not metabolically
stable in vivo [31, 32]. Our studies also confirmed that LT-106-175 was not stable in
human liver microsomes (t1/2: 51 min). Additionally, previous introduction of
isopropyl to thiophene ring showed only a little influence on the kinase and cell
activity (compound 40 and 41). Hence, we chose to replace the thiophene ring
directly with other heterocycles (data not shown) or saturated cycles. Among those
compounds, compound 50 was obtained by replacing the thiophene (44) with
cyclopentane ring, and it showed the highest inhibitory activity against FLT3 (IC50:
0.213 nM). In addition, its selectivity for FLT3 also increased obviously when
compared with compound 44. Furthermore, compound 50 also had significant
anti-proliferation activity in MV4-11 (IC50: 0.016 µM). The similar replacement of
thiophene with cyclohexane ring (51) did not affect the cell activity of compounds,
but caused a slight decrease in FLT3 inhibitory activity as well as the selectivity for
FLT3 (Table 3). In all, compound 50 was characterized to be the most potent and

ACCEPTED MANUSCRIPT
selective FLT3 inhibitor in this study, with the remarkable anti-growth activity against
MV4-11.
Table 3. Optimization of the thiophene ring in the hydrophobic pocket
(% enzyme activity) a
/IC50 (nM) b
( % cell growth inhibition) c
/IC50 (µM) d
Comp
ounds
R1
R2
CDK2
CDK6
FLT3
MV4-11
0.21 ±
0.05
86.30%
79.38%
91.25%
0.016 ± 0.006
0.018 ± 0.003
50
51
0.66 ±
0.03
87.22%
a
b
% Enzyme activity at 0.123 µM in the presence of 10 µM ATP; In the presence of 10 µM
ATP with 0.123 µM compound, the values are the mean ± SD from three independent experiments.
c % cell growth inhibition at 1 µΜ; d the highest test contraction is 1 µM, the values are the mean ±
SD from three independent experiments.
Pharmacokinetics of compound 50 in Human Liver Microsomes
In order to verify whether the metabolic stability of compound 50 was improved,
the human liver microsomes (HML) assay was performed. As data shown (Table 4),
compound 50 was metabolically more stable (t1/2: 147 min) than the lead compound
LT-106-175 (t1/2: 51 min).
Table 4. Pharmacokinetics of compound 50 in human liver microsomes.
Human Liver Microsomes

ACCEPTED MANUSCRIPT
T1/2
(min)
51
CLint
(µL/min/mg protein)
13.58
EH
Compound
(%)
29.7
28.7
LT-106-175
50
147
9.40
a T1/2 is half life. b CLint is the intrinsic clearance. c liver uptake rate.
Kinase profiling
Through the structure-based optimization process, a relatively selective FLT3
inhibitor compound 50 was identified using a FLT3/CDK2/6 multiple kinase inhibitor
as lead. Then, the selectivity profile of compound 50 was further characterized by the
kinome assay on a panel of kinases. Thirty one kinases that were efficiently inhibited
by LT-106-175 were selected (data have not been published). Generally, compound 50
displayed the highest inhibitory activity against FLT3 (98%) and little activity against
CDKs. More importantly, compound 50 exhibited no inhibitory activity against c-Kit,
which was usually vulnerable to other FLT3 inhibitors, such as PKC412 and
AC220[33, 34]. Simultaneous inhibition of both FLT3 and c-Kit correlated with the
myelosuppression toxicity observed from the most advanced clinical trials[23, 35].
Subsequently, kinases whose inhibitory rates exceed 50% were selected for IC50
determination. As shown in Table 5, compound 50 still showed excellent inhibitory
activity against FLT3, the IC50 value of which was almost 120-fold stronger than that
of FLT4/VEGFR3 (IC50: 23.01 nM). Only weak inhibitory activity against
FLT1/VEGFR1 (IC50: 220.19 nM) and KDR/VEGFR2 (IC50: 94.62 nM) was found. In
comparison, LT-106-175 exhibit excellent inhibitory activity against CDK2/Cyclin A1,

ACCEPTED MANUSCRIPT
CDK4/cyclinD1, CDK6/cyclinD1, FLT1/VEGFR1, ERK7/MAPK15, FLT4/VEGFR3
(IC50 < 10 nM) in addition to FLT3. All the data above confirmed that compound 50
was highly selective against FLT3 (Figure 2).
Figure 2. Kinase inhibition profile of compound 50 against 31 kinases.
Table 5. FLT3 selective inhibition by compound 50
Compound 50
(nM) a
LT-106-175
Staurosporine
IC50 (M) a
Kinase
0.19 ± 0.04
220.19 ± 1.50
94.62 ± 3.14
23.01 ± 2.31
0.60 ± 0.04
19.8 ± 0.43
35.1 ± 1.56
4.5 ± 0.32
1.52 ± 0.21
15.97 ± 1.26
1.67 ± 0.11
3.27 ± 0.36
FLT3
FLT1/VEGFR1
KDR/VEGFR2
FLT4/VEGFR3
a
In the presence of 10 µM ATP, the IC50 are the mean ± SD from three independent
experiments proceed by reaction biology company
Considering FLT3 mutation is very common and often indicates poor outcomes,

ACCEPTED MANUSCRIPT
the efficacy of compound 50 against a variety of FLT3-ITD/TKD mutations was
tested. As data shown in Table 6, compound 50 showed the significant inhibitory
activity against those FLT3 mutants, even some of which were clinically identified as
quizartinib-resistant, such as FLT3-D835Y, FLT3 (ITD)-F691L and FLT3
(ITD)-D835V[36, 37], etc. This indicated that compound 50 may be of promise to
overcome the currently known drug resistances. It should be noted that lead
compound LT-106-175 showed similar activity against those FLT3 mutants,
indicating that the structural optimization did not affect FLT3 inhibitory potency
eventually.
Table 6. The inhibitory activity of compound 50 against a panel of FLT3 mutants
Compound 50 Staurosporin
LT-106-175
IC50 / Kd (nM)
0.54 ± 0.03 a
1.23 ± 0.10 a
7.35 ± 0.81 a
17.54 ± 2.43 a
1.85 ± 0.34 a
2.67 ± 0.31 a
13.22 ± 1.15 a
7.01 ± 0.52 b
0.23 ± 0.11 b
1.30 ± 0.002 b
Kinase
IC50 / Kd (nM)
1.79 ± 0.04 a
6.97 ± 0.08 a
9.15 ± 0.69 a
14.94 ± 1.51 a
12.85 ± 1.20 a
8.74 ± 0.77 a
15.28 ± 1.24 a
5.33 ± 0.49 b
3.70 ± 0.27 b
13.12 ± 0.92 b
IC50 (nM)
0.71 ± 0.09
1.74 ± 0.17
2.03 ± 0.26
5.49 ± 0.73
1.96 ± 0.25
0.83 ± 0.09
4.95 ± 0.71
ndc
FLT3 (D835Y)
FLT3 (F594_R595insR)
FLT3 (F594_R595insREY)
FLT3 (ITD)-NPOS
FLT3 (ITD)-W51
FLT3 (R595_E596)
FLT3 (Y591-V592)
FLT3(D835Y)
ndc
FLT3 (ITD)-D835V
FLT3 (ITD)-F691L
ndc

ACCEPTED MANUSCRIPT
a
In the presence of 10 µM ATP, the IC50 are the mean ± SD from three independent
experiments proceed by reaction biology company. b Kd value provided by DiscoverX.
c not tested
Molecular Modeling of Compound 50 with FLT3
Figure 3. Molecular docking analysis of compound 50 bound to FLT3. (A) Compound 50 docked
into homology model of FLT3. (B) Overlapping of the docked poses of compound 50 (Light blue)
and LT-106-175 (yellow) in FLT3.
As one of the most selective FLT3 inhibitors in this series, compound 50 was
docked into the homology model of FLT3 in order to explain the observed SAR. As
shown in Figure 3, compound 50 adopted a similar binding mode with FLT3 as
LT-106-175. The pyrazole ring bond to the kinase hinge region via two hydrogen
bonds. In the hydrophilic pocket, the piperazine ring formed a hydrogen bond with
Asn701, which helped improve the FLT3 inhibitory potency than other hydrophilic
groups. The 6,7-dihydro-5H-cyclopenta[d]pyrimidine of compound 50 occupy the
hydrophobic cavity. At the back of the ATP binding site the 2-methyl group made
favorable van der waals interactions with the gatekeeper residue Phe691 around the
back pocket. This further explains the result that introduction of methyl is the key
factor to improve the selectivity for FLT3.

ACCEPTED MANUSCRIPT
Cell activity
In assessing the anti-proliferative effect of compound 50 in other leukemia cell
lines and lymphocytes cell lines, K562, MV4-11, MOLT-4, HL-60, MOLM-13 and
Jurkat cell lines were selected for evaluation (Table 7). Consistent with its FLT3
inhibitory activity, FLT3-ITD-expressing cells exhibited extreme sensitivity to
compound 50. Compound 50 showed the most potent anti-proliferative activity
against MV4-11 and MOLM-13 cells, with the IC50 value of 0.016 and 0.037 µM,
respectively. Moreover, compound 50 had no efficacy in FLT3-low or FLT3–no
expression cells,
indicating that compound 50 may be less toxic against
FLT3-independent cells in vitro.
Table 7. Effects of compound 50 on cell lines with different FLT3-expressing status
IC50 (µM)a
Cell line
MV4-11 MOLM-13 MOLT-4
HL-60
K562
Jurkat
ITD
ITD
not known
WT/lowb FLT3-nullc FLT3-null
FLT3 status
0.016 ±
0.005
0.037 ±
0.011
2.115 ±
0.981
26.690 ± 11.62 ±
4.595 ±
0.653
Compound 50
LT-106-175
2.547
1.280
0.016±
0.006
0.039 ±
0.002
NT
0.334d
0.235 d
NT
a The highest test contraction is 1 µM, the values are the mean ± SD from three independent
b
c
experiments. Cell line express less wide type FLT3. cell lines not driven by oncogenic FLT3.
d GI50 value were provided by NCI
Mechanism study

ACCEPTED MANUSCRIPT
In order to further investigate whether the anti-AML activity is associated with
the FLT3 inhibition and its downstream signaling proteins, we examined the effects of
compound 50 on the phosphorylation of FLT3, STAT5, ERK and AKT in MV4-11
cells (Figure 4). As shown by Western blot assays, compound 50 could decrease the
phosphorylation of FLT3 and its downstream signaling proteins, i.e. AKT, ERK and
STAT5, in a dose-dependent manner. However, the down-regulatory effect of
compound 50 on p-FLT3 at the concentration was not as strong as quizartinib
(AC220). This may be due to relatively weak cell potency of compound 50 compared
with quizartinib[38]. In conclusion, the anti-growth activity of compound 50 against
MV4-11 was closely related to the inhibition of FLT3 signal pathway.
Figure 4. MV4-11 cells were treated with compound 50 or AC220 for 4 h, and the
phosphorylation of FLT3, STAT5, ERK and AKT protein was analyzed by immune blotting.
In MV4-11 cell line, compound 50 induced cell apoptosis as detected by annexin
V staining. A dose-dependent increase in the percentage of apoptotic cells was
observed for compound 50, which was consistent with AC220 (Figure 5).

ACCEPTED MANUSCRIPT
Figure 5. MV4-11 cells were treated with compound 50 for 24 h and then analyzed by Annexin
V/PI staining. The percentage of cells undergoing apoptosis was defined as the sum of early
apoptosis, advanced apoptosis and necrotic cells. (A) DMSO control; (B) compound 50 at 0.2 µM;
(C) compound 50 at 0.5 µM; (D) compound 50 at 1 µM; (E) AC220 at 1 µM; (F) Percentage of
apoptosis cells.
Conclusion
In summary, from the modification of a CDKs/FLT3 multi-kinase inhibitor
LT-106-175 using the structure-guided drug design approach, a series of
1H-pyrazole-3-carboxamide derivatives were synthesized with the aim of improving
the FLT3 selectivity. As one of the most potent mutated FLT3 inhibitor, compound 50
showed great selectivity for FLT3, especially without the inhibitory activity against
c-Kit. Compound 50 also exhibited significant inhibitory efficacy to multiple FLT3
mutants, and thus it was expected to overcome the acquired resistance due to those
mutations. Potent and selective growth inhibition activities were observed for
compound 50 against the leukemia cell lines harboring FLT3-ITD mutant, such as

ACCEPTED MANUSCRIPT
MV4-11 and MOLM-13, in vitro. We also preliminarily clarified that compound 50
indeed interfered FLT3 signal transduction in MV4-11 cell, and it induced apoptosis
in a dose-dependent manner. Combining its selective inhibition of FLT3 mutants and
FLT3-ITD expressing cells as well as the results from mechanism studies, it can be
concluded that compound 50 may be a potential drug candidate for the treatment of
AML.
EXPERMENTAL SECTION
General Procedures
Unless otherwise specified, reagents were purchased from commercial suppliers and
used without further purification. Melting points were determined by X-4 digital
display micro-melting point apparatus (Beijing Tech Instrument Co., Ltd.); NMR
1
spectra were recorded on Bruker AVANCE AV-800 spectrometer (800 MHz for H,
150 MHz for 13C), Bruker AVANCE AV-600 spectrometer (600 MHz for 1H, 126 MHz
for 13C) or Bruker AVANCE AV-300 spectrometer (300 MHz for 1H, 75 MHz for 13C);
Mass spectra were obtained on the Agilent 1100 LC / MSD mass spectrometer
(Agilent, USA) and Q-tofmicro MS (micromass company). All reactions were
monitored by TLC (Merck Kieselgel GF254) and spots were visualized with UV light
or iodine. The purity of biologically evaluated compounds was >95% as determined
by HPLC.
General procedure A for the synthesis of compounds 9-12
4-Nitrobenzyl bromide (46.3 mmol) was dissolved in dichloromethane (100 mL).
The solution was added to the mixture of relative amine (47.0 mmol) and

ACCEPTED MANUSCRIPT
triethylamine (70.3 mmol) in dichloromethane (20 ml). The reaction mixture was
stirred at r.t. for 24 h and was extracted with dichloromethane (100 ml × 3). After
removal of the solvent, the residue was crystallized from ethanol, giving yellow
powder. Compounds 1 and 2 were used for further reaction without purification.
To a suspension of compounds 1-2 (36.2 mmol) in 95% ethanol (100 ml), 85%
.
NH2NH2 H2O (362 mmol), 95% ethanol (100 ml) and iron (III) oxide hydroxide
(FeO(OH)/C, 2.0 g) were added and heated to reflux. When TLC analysis showed
complete conversion of the starting material, the reaction mixture was filtrate through
Cellit and the filtrate was concentrated in vacuum. The crude product was purified by
silica gel colum chromatography (DCM/MeOH) to yield the title compound (3 and 4)
as white solid.
The mixture of compound 4 (1eq, 18.5 mmol), 4-Nitro-1H-pyrazole-3-acid
(1.1equiv, 20.4mmol), EDC (1.2equiv, 22.2mmol), HOBT (1.2equiv, 22.2mmol) in
DMF (50 ml) was stirred for 24 hours. The ice water (100 ml) was added to the
reaction mixture. A large amount of yellow solid precipitation (compound 8) was
acquired. Compound 8 was used without further purification. Compounds 8 was
reduced by the same process as compound 4, and then the resulting compound 12 was
purified by column chromatography on silica gel, eluted with the appropriate solvent.
General procedure B for the synthesis of compounds 22-24
4-Fluoronitrobenzene (46.3 mmol) and K2CO3 were dissolved in DMSO (50 mL).
The reaction mixture was stirred at r.t. for 30 min and then tert-butyl
piperazine-1-carboxylate (69.5 mmol) was added. The reaction mixture was stirred at

ACCEPTED MANUSCRIPT
70℃ for 5h. The ice water (500 ml) was added to the reaction mixture. A large
amount of yellow solid precipitation (compound 13) was acquired. Compound 13 was
used for further reaction without purification. Compound 22 was synthesized by
following the same synthetic process as the synthesis of compound 12.
General procedure C for the synthesis of compounds 25-43
Compounds 9-12 or 22-24 (1 equiv.) and corresponding chorides (1.2 equiv.) in
AcOH/H2O:1/1 (20 ml) were heated to 50 °C. When TLC analysis showed the
complete conversion of the starting material, sodium hydroxide (2 equiv.) was added
to the mixture. Then, the reaction mixture was extracted with ethyl acetate. After
removal of the solvent, compounds 33-43 were purified by column chromatography
and the corresponding compounds were obtained. Compounds 25-32 were used
without further purification.
General procedure D for the synthesis of compounds 44-51
Compound 25-32 (1 equiv.) were dissolved in dichloromethane/CF3COOH and
stirred at r.t. for 5 h. When TLC analysis showed complete conversion of the starting
material, sodium hydroxide (2 equiv.) was added to the mixture. Then, the reaction
mixture was extracted with dichloromethane. After removal of the solvent, the residue
was purified by column chromatography, and the corresponding compounds were
obtained.
4-((7H-pyrrolo[2,3-d]pyrimidin-4-yl)amino)-1-methyl-N-(4-((4-methylpiperazin-
1-yl)methyl)phenyl)-1H-pyrazole-3-carboxamide (33). Compound 33 was prepared
according to general procedure C on 0.5 mmol scale. Purification by column

ACCEPTED MANUSCRIPT
chromatography (3% MeOH/DCM) yield the title compound (98 mg, 0.22 mmol
Yield 44%). white solid; m.p 238-240 °C. HPLC analysis: retention time = 6.954 min;
1
peak area, 98.55%. H-NMR (300 MHz, DMSO-d6) δ 11.95 (s, 1H), 10.28 (s, 1H),
9.58 (s, 1H), 8.60 (s, 1H), 8.39 (s, 1H), 7.79 (d, J = 8.1 Hz, 2H), 7.33 (s, 1H), 7.25 (d,
J = 8.1 Hz, 2H), 6.49 (s, 1H), 4.03 (s, 3H), 3.42 (s, 2H), 2.36 (s, 6H), 2.17 (s, 3H).
13C-NMR (126 MHz, DMSO-d6) δ 162.82 (s), 151.94 (s), 151.61 (s), 151.09 (s),
137.51 (s), 134.05 (s), 131.99 (s), 129.53 (s), 125.73 (s), 123.64 (s), 123.23 (s),
121.14 (s), 103.59 (s), 97.18 (s), 62.10 (s), 55.10 (s), 52.80 (s), 46.03 (s). HRMS-EI
m/z [M+H] + calcd for C23H28N9O: 446.2417, found: 446.2386.
4-((7H-pyrrolo[2,3-d]pyrimidin-4-yl)amino)-5-methyl-N-(4-((4-methylpiperazin-
1-yl)methyl)phenyl)-1H-pyrazole-3-carboxamide (34). Compound 34 was prepared
according to general procedure C on 0.5 mmol scale. Purification by column
chromatography (3% MeOH/DCM) yield the title compound (103 mg, 0.23 mmol
Yield 46%). white solid; m.p 175-177 °C. HPLC analysis: retention time = 4.183 min;
1
peak area, 98.77%. H-NMR (300 MHz, DMSO-d6) δ 11.65 (s, 1H), 10.02 (s, 1H),
8.73 (s, 1H), 8.07 (s, 1H), 7.63 (d, J = 7.9 Hz, 2H), 7.16 (m, 2H), 6.45 (s, 1H), 3.36 (s,
2H), 2.29 (s, 8H), 2.14 (d, J = 8.7 Hz, 6H). 13C-NMR (151 MHz, DMSO-d6) δ 155.99
(s), 151.56 (s), 151.31 (s), 138.01 (s), 133.59 (s), 129.51 (s), 122.06 (s), 120.04 (s),
119.16 (s), 103.38 (s), 99.25 (s), 62.11 (s), 55.17 (s), 52.93 (s), 46.19 (s). HRMS-EI
m/z [M+H] + calcd for C23H28N9O: 446.2417, found: 446.2383.
5-methyl-N-(4-((4-methylpiperazin-1-yl)methyl)phenyl)-4-(thieno[2,3-d]pyrimidi
n-4-ylamino)-1H-pyrazole-3-carboxamide (35). Compound 35 was prepared

ACCEPTED MANUSCRIPT
according to general procedure C on 0.5 mmol scale. Purification by column
chromatography (3% MeOH/DCM) yield the title compound (118 mg, 0.26 mmol,
Yield 51%). white solid; m.p 92-94 °C. HPLC analysis: retention time = 4.961 min;
1
peak area, 98.42%. H-NMR (300 MHz, DMSO-d6) δ 13.27 (s, 1H), 10.00 (s, 1H),
9.29 (s, 1H), 8.29 (s, 1H), 7.77 (s, 1H), 7.66 (d, J = 6.4 Hz, 3H), 7.17 (d, J = 8.2 Hz,
2H), 3.37 (s, 2H), 2.33 (s, 8H), 2.17 (d, J = 3.2 Hz, 6H). 13C-NMR (151 MHz,
DMSO-d6) δ 166.68 (s), 157.09 (s), 154.02 (s), 138.04 (s), 133.45 (s), 129.49 (s),
123.66 (s), 120.20 (s), 120.09 (s), 118.22 (s), 116.97 (s), 62.04 (s), 55.02 (s), 52.72 (s),
45.97 (s), 21.61 (s). HRMS-EI m/z [M+H] + calcd for C23H27N8OS: 463.2029, found:
463.2024.
N-(4-((4-methylpiperazin-1-yl)methyl)phenyl)-4-((2-methylthieno[2,3-d]pyrimidi
n-4-yl)amino)-1H-pyrazole-3-carboxamide (36). Compound 36 was prepared
according to general procedure C on 0.5 mmol scale. Purification by column
chromatography (5% MeOH/DCM) yield the title compound (112 mg, 0.24 mmol
Yield 48%). white solid; m.p 270-272 °C. HPLC analysis: retention time = 3.990 min;
1
peak area, 99.44%. H-NMR (300 MHz, DMSO-d6) δ 13.53 (s, 1H), 10.35 (s, 1H),
9.94 (s, 1H), 8.58 (s, 1H), 7.79 (d, J = 7.7 Hz, 2H), 7.67 (d, J = 5.8 Hz, 1H), 7.41 (d, J
= 5.7 Hz, 1H), 7.26 (d, J = 8.1 Hz, 2H), 3.42 (s, 2H), 2.61 (s, 3H), 2.36 (s, 8H), 2.17
13
(s, 3H). C-NMR (151 MHz, DMSO-d6) δ 167.30 (s), 163.10 (s), 153.09 (s), 137.45
(s), 134.14 (s), 129.53 (s), 124.18 (s), 124.14 (s), 121.14 (s), 117.79 (s), 114.54 (s),
+
62.11 (s), 55.12 (s), 52.85 (s), 46.09 (s), 26.20 (s). HRMS-EI m/z [M+H] calcd for
C23H27N8OS: 463.2029, found: 163.1995.

ACCEPTED MANUSCRIPT
4-((2-methylthieno[2,3-d]pyrimidin-4-yl)amino)-N-(4-(morpholinomethyl)phenyl
)-1H-pyrazole-3-carboxamide (37). Compound 37 was prepared according to
general procedure C on 0.5 mmol scale. Purification by column chromatography (5%
MeOH/DCM) yield the title compound (117 mg, 0.26 mmol Yield 52%). white solid;
m.p 263-264 °C. HPLC analysis: retention time = 6.368 min; peak area, 98.44%.
1H-NMR (300 MHz, DMSO-d6) δ 13.51 (s, 1H), 10.37 (s, 1H), 9.93 (s, 1H), 8.59 (s,
1H), 7.80 (d, J = 8.3 Hz, 2H), 7.68 (d, J = 5.9 Hz, 1H), 7.41 (d, J = 6.0 Hz, 1H), 7.28
13
(d, J = 8.4 Hz, 2H), 3.58 (s, 4H), 3.43 (s, 2H), 2.61 (s, 3H), 2.35 (s, 4H). C-NMR
(151 MHz, DMSO-d6) δ 167.30 (s), 163.10 (s), 153.07 (s), 137.53 (s), 133.68 (s),
133.37 (s), 129.63 (s), 124.17 (s), 124.14 (s), 121.26 (s), 121.15 (s), 117.78 (s),
114.54 (s), 66.67 (s), 62.52 (s), 53.61 (s), 26.20 (s). HRMS-EI m/z [M+H] + calcd for
C22H24N7O2S: 450.1712, found: 450.1673.
4-((2-cyclopropylthieno[2,3-d]pyrimidin-4-yl)amino)-N-(4-((4-methylpiperazin-1-
yl)methyl)phenyl)-1H-pyrazole-3-carboxamide (38). Compound 38 was prepared
according to general procedure C on 0.5 mmol scale. Purification by column
chromatography (5% MeOH/DCM) yield the title compound (95 mg, 0.20 mmol
Yield 39%). yellow solid; m.p 138-139 °C. HPLC analysis: retention time = 7.091
1
min; peak area, 98.85%. H-NMR (300 MHz, DMSO-d6) δ 13.53 (s, 1H), 10.37 (s,
1H), 9.92 (s, 1H), 8.43 (s, 1H), 7.78 (d, J = 8.2 Hz, 2H), 7.62 (d, J = 5.9 Hz, 1H), 7.38
(d, J = 6.0 Hz, 1H), 7.26 (d, J = 8.2 Hz, 2H), 3.42 (s, 2H), 2.34 (s, 8H), 2.21 (s, 1H),
2.14 (s, 3H), 1.06 (m, 4H). 13C-NMR (151 MHz, DMSO-d6) δ 175.31 (s), 167.42 (s),
163.24 (s), 153.11 (s), 137.53 (s), 133.76 (s), 133.39 (s), 129.55 (s), 124.21 (s),

ACCEPTED MANUSCRIPT
123.62 (s), 121.11 (s), 120.99 (s), 117.69 (s), 114.82 (s), 55.19 (s), 52.97 (s), 46.22 (s),
+
18.55 (s), 10.30 (s). HRMS-EI m/z [M+H] calcd for C25H29N8OS: 489.2185, found:
489.2165.
4-((2-cyclopropylthieno[2,3-d]pyrimidin-4-yl)amino)-N-(4-(morpholinomethyl)p
henyl)-1H-pyrazole-3-carboxamide (39). Compound 39 was prepared according to
general procedure C on 0.5 mmol scale. Purification by column chromatography (5%
MeOH/DCM) yield the title compound (100 mg, 0.21 mmol Yield 41%). white solid;
m.p 253-255 °C. HPLC analysis: retention time = 7.057 min; peak area, 100.00%.
1H-NMR (300 MHz, DMSO-d6) δ 13.51 (s, 1H), 10.36 (s, 1H), 9.92 (s, 1H), 8.44 (s,
1H), 7.80 (d, J = 7.9 Hz, 2H), 7.62 (d, J = 5.8 Hz, 1H), 7.37 (d, J = 6.0 Hz, 1H), 7.28
(d, J = 7.9 Hz, 2H), 3.57 (s, 4H), 3.43 (s, 2H), 2.35 (s, 4H), 2.21 (s, 1H), 1.07 (m, 4H).
13C-NMR (126 MHz, DMSO-d6) δ 167.43 (s), 167.36 (s), 163.14 (s), 153.09 (s),
137.52 (s), 133.74 (s), 133.31 (s), 129.65 (s), 124.26 (s), 123.60 (s), 121.17 (s),
120.75 (s), 117.78 (s), 114.73 (s), 66.68 (s), 62.53 (s), 53.61 (s), 18.54 (s), 10.38 (s).
HRMS-EI m/z [M+H] + calcd for C24H26N7O2S: 476.1869, found: 476.1826.
4-((6-isopropyl-2-methylthieno[2,3-d]pyrimidin-4-yl)amino)-N-(4-(morpholinom
ethyl)phenyl)-1H-pyrazole-3-carboxamide (40). Compound 40 was prepared
according to general procedure C on 0.5 mmol scale. Purification by column
chromatography (5% MeOH/DCM) yield the title compound (135 mg, 0.28 mmol
Yield 55%). white solid; m.p 263-264 °C. HPLC analysis: retention time = 3.750 min;
1
peak area, 98.43%. H-NMR (300 MHz, DMSO-d6) δ 13.50 (s, 1H), 10.33 (s, 1H),
9.75 (s, 1H), 8.57 (s, 1H), 7.79 (d, J = 7.9 Hz, 2H), 7.28 (d, J = 8.1 Hz, 2H), 7.07 (s,

ACCEPTED MANUSCRIPT
1H), 3.57 (s, 4H), 3.42 (s, 2H), 3.30 (m, 1H), 2.58 (s, 3H), 2.34 (s, 4H), 1.34 (d, J =
6.8 Hz, 6H). 13C-NMR (151 MHz, DMSO-d6) δ 163.15 (s), 162.34 (s), 152.27 (s),
150.04 (s), 137.50 (s), 133.68 (s), 133.17 (s), 129.61 (s), 124.34 (s), 121.23 (s),
114.54 (s), 111.94 (s), 66.67 (s), 62.52 (s), 53.60 (s), 30.51 (s), 26.15 (s), 24.61 (s).
HRMS-EI m/z [M+H] + calcd for C25H30N7O2S: 492.2182, found: 492.2029..
4-((6-isopropyl-2-methylthieno[2,3-d]pyrimidin-4-yl)amino)-N-(4-((4-methylpipe
razin-1-yl)methyl)phenyl)-1H-pyrazole-3-carboxamide (41). Compound 41 was
prepared according to general procedure C on 0.5 mmol scale. Purification by column
chromatography (5% MeOH/DCM) yield the title compound (115 mg, 0.23 mmol
Yield 45%). white solid; m.p 151-153 °C. HPLC analysis: retention time = 7.214 min;
1
peak area, 98.39%. H-NMR MHz, (300 MHz, DMSO-d6) δ 13.50 (s, 1H), 10.33 (s,
1H), 9.75 (s, 1H), 8.57 (s, 1H), 7.79 (d, J = 7.9 Hz, 2H), 7.28 (d, J = 8.1 Hz, 2H), 7.07
(s, 1H), 3.57 (s, 4H), 3.42 (s, 2H), 3.30 (m, 1H), 2.58 (s, 3H), 2.34 (s, 4H), 1.34 (d, J
13
= 6.8 Hz, 6H). C-NMR (151 MHz, DMSO-d6) δ 166.03 (s), 162.34 (s), 152.31 (s),
150.02 (s), 137.76 (s), 129.77 (s), 124.36 (s), 121.21 (s), 114.53 (s), 111.98 (s), 61.31
(s), 53.69 (s), 50.80 (s), 43.98 (s), 30.51 (s), 26.16 (s), 24.61 (s). HRMS-EI m/z [M+H]
+ calcd for C26H33N8OS: 505.2498, found: 505.2463.
N-(4-(4-methylpiperazin-1-yl)phenyl)-4-((2-methylthieno[2,3-d]pyrimidin-4-yl)a
mino)-1H-pyrazole-3-carboxamide (42). Compound 42 was prepared according to
general procedure C on 0.5 mmol scale. Purification by column chromatography (5%
MeOH/DCM) yield the title compound (112 mg, 0.25 mmol Yield 49%). yellow solid;
m.p >269 °C. HPLC analysis: retention time = 6.587 min; peak area, 98.83%.