Synthesis and biological evaluation of benzamides and benzamidines as selective inhibitors of VEGFR tyrosine kinases

Article abstract of DOI:10.1016/j.bmcl.2006.07.075

A series of benzamidines and benzamides was synthesized as selective inhibitors of vascular endothelial growth factor receptor (VEGFR) tyrosine kinases, and tested for inhibitory activity toward autophosphorylation by the enzyme assay. Selective inhibitio

Full text of DOI:10.1016/j.bmcl.2006.07.075

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5127–5131
Synthesis and biological evaluation of benzamides and benzamidines
as selective inhibitors of VEGFR tyrosine kinases
Hiroyuki Nakamura,a,* Yusuke Sasaki,a Masaharu Uno,a Tomohiro Yoshikawa,a
Toru Asano,a Hyun Seung Ban,a Hidesuke Fukazawa,b
Masabumi Shibuyac and Yoshimasa Ueharab
aDepartment of Chemistry, Faculty of Science, Gakushuin University, Tokyo 171-8588, Japan
bDepartment of Bioactive Molecules, National Institute of Infectious Diseases, Tokyo 162-8640, Japan
cInstitute of Medical Science, University of Tokyo, Tokyo 108-8639, Japan
Received 23 March 2006; revised 26 June 2006; accepted 11 July 2006
Available online 8 August 2006
Abstract—A series of benzamidines and benzamides was synthesized as selective inhibitors of vascular endothelial growth factor
receptor (VEGFR) tyrosine kinases, and tested for inhibitory activity toward autophosphorylation by the enzyme assay. Selective
inhibition of VEGFR-2 tyrosine kinase was observed in the salicylic amide 4e and the anthranilic amidine 5a, and their percent inhi-
bitions of VEGFR-2 tyrosine kinase were 44–60% at a 10 lM concentration of compounds. The salicylic amide 4a showed inhibition
of both VEGFR-1 and VEGFR-2 tyrosine kinases at a 10 lM concentration.
Ó 2006 Elsevier Ltd. All rights reserved.
Angiogenesis is tightly regulated and only occurs nor-
mally in inflammation, wound healing, and the female
reproductive cycle in the adult.1 Uncontrolled angio-
genesis involves pathological states such as atheroscle-
rosis, diabetic retinopathy, rheumatoid arthritis, and
solid tumor growth. Vascular endothelial growth fac-
tor (VEGF) is a key growth factor in tumor angiogen-
esis, therefore, its receptor tyrosine kinases known as
VEGFR-1 (Flt-1) and VEGFR-2 (KDR/Flk-1) have
been considered as attractive targets for development
of anti-cancer drugs.2 Various small molecule VEGFR
kinase inhibitors have been developed, such as
SU6668,3 ZD6474,4 PTK787,5 AAL993,6 the pyra-
zine-pyridine compound 1,7 and aminopyrimidine 28
(Fig. 1). We focused on the anthranilic amide frame-
work of AAL993. According to the X-ray structural
analysis, AAL993 interacted with VEGFR2 kinase
domain through hydrogen bonds between the pyri-
dine-N and the backbone-NH of Cys919 in the hinge
region of the kinase, the anthranilamide–C@O and the
backbone-NH of Asp1046 of the DFG-loop, and the
anthranilamide-NH with the side-chain carboxylate
of Glu885 of helix-C as well as the pseudocycle for-
mation by an intramolecular hydrogen bond between
the aniline–NH and the benzanilide–C@O.9 We are
interested in an amidine skeleton as a mimic of an
amide. Amidines are, in general, more basic in com-
parison with ammonia, therefore expected different
properties from the corresponding amides. Previously,
we synthesized 4,5-dimethoxyanthranilic amides and
4,5-dimethoxyanthranilic amidines as a mimic of the
4-anilinoquinazoline framework for inhibition of
epidermal growth factor receptor (EGFR) tyrosine
kinase.10 In this paper, we synthesized a series of the
benzamides (3 and 4) and benzamidines (5 and 6)
(Fig. 2), and evaluated their inhibitory activity of
VEGFR1/2 tyrosine kinases.
Anthranilic amides 3a–f were synthesized from methyl
2-aminobenzoate 7a and methyl 4,5-dimethoxy-2-ami-
nobenzoate 7b as shown in Scheme 1. Reductive
amination reactions of 7a and 7b with 4-pyridinecarbox-
aldehyde gave the corresponding anthranilic acid ester
8a and 8b in 69% and 79% yields, respectively. Amide
formation of the esters 8 with various amines was
promoted by trimethylaluminum to afford the
Keywords: VEGFR-1; VEGFR-2; Benzamides; Benzamidines;
VEGFR tyrosine kinases; Inhibitor.
*
Corresponding author. Tel.: +81 339860221; fax: +81 359921029;
e-mail: hiroyuki.nakamura@gakushuin.ac.jp
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.07.075

5128
H. Nakamura et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5127–5131
Cl
CO2H
F
Br
HN
HN
N
N
N
O
O
H
N
O
N
N
H
N
N
PTK 787
ZD 6474
SU 6668
H
O
N
Cl
N
HN
O
CF3
H
H
N
CF3
N
N
O
H
N
F
O
N
N
H
OH
NH2
N
N
N
AAL993
1
2
Figure 1. Structures of VEGFR tyrosine kinase inhibitors.
R2
from 2-aminobenzonitrile 12a or 4,5-dimethoxyl-2-amino-
benzonitrile 12b (Scheme 4). Reductive amination of 12a–
b with 4-pyridinecarboxaldehyde proceeded in the pres-
ence of NaCNBH3, giving the corresponding benzonitr-
iles 13a and 13b in 41% and 92% yields, respectively.
Amidine formation of 13a–b with various amines was
promoted by trimethylaluminum to afford the corre-
sponding anthranilic amidines 5a–f in 30–71% yields (Ta-
ble 1). Salicylic amidines 6a–f were also synthesized from
2-hydroxybenzonitrile14a and 4,5-dimethoxyl-2-hydroxy-
benzonitrile 14b along with Scheme 5. In the amidine syn-
thesis, we examined the reaction of benzonitriles, such as
12b and 13b, with pyrazin-2-amine and/or pyrimidin-2-
amine in order to compare with 3e and 3f, however, the
desired amidines were not obtained.
HN
R1
R1
X = O, Y = NH: anthranilic amides, (3)
X
X = O, Y = O: salicylic amides, (4)
X = NH, Y = NH: anthranilic amidines, (5)
X = NH, Y = O: salicylic amidines, (6)
Y
R3
Figure 2. Design of benzamides and benzamidines as inhibitors of
VEGFR tyrosine kinases.
corresponding anthranilic amides 3a–f in 30–71% yields
as shown in Table 1.
Salicylic amides 4a–f were synthesized from 2-methoxy-
benzoic acids 9a (R1 = H) or 9b (R1 = MeO) as shown in
Scheme 2. The 2-methoxybenzoic acids 9a–b were treat-
ed with thionyl chloride, and the resulting acid chlorides
reacted with various amines to afford the 2-methoxy
benzoic acid amides in 50–99% yields, which underwent
selective demethylation at C-2 position in the presence
of boron trichloride,11 giving the salicylic amides 10a–f
in 23–78% yields. The reaction of 10a–f with 4-chlorom-
ethylpyridine gave the salicylic amides 4a–f.12
Inhibitory activity of the anthranilic amides 3a–f, salicylic
amides 4a–h, anthranilic amidines 5a–f, and salicylic ami-
dines 6a–f was investigated by ELISA using VEGFR1
(Flt-1) and VEGFR2 (KDR).13,14 Percents of kinase inhi-
bition at a 10 lM concentration of compounds are shown
in Table 1. Among anthranilic amides 3a–f, 3b showed
12% and 30% kinase inhibitions of Flt-1 and KDR,
respectively. Salicylic amide 4a showed significant inhibi-
tion of both Flt-1 and KDR (50% and 60%, respectively),
whereas, 4b, 4e, and 4h showed selective inhibition of
KDR (24%, 55%, and 24%, respectively). No kinase inhi-
bition was observed in a series of benzamidines 5 and 6,
except for 5a, which showed 44% kinase inhibition of
KDR. We also examined EGFR tyrosine kinase assay,
however, no inhibition was observed in compounds 3–6
at a 10 lM concentration.
Scheme 3 shows the synthesis of the salicylic amides
4g–h, which were not obtained by the above synthetic
scheme. Treatment of 9b with BCl3 gave the methyl ester
11, which underwent ether formation with 4-chloro-
methylpyridine followed by amide formation with vari-
ous amines using AlMe3 to give the salicylic amides 4g–h.
We next synthesized the anthranilic and salicylic amidines
5 and 6. The anthranilic amidines 5a–f were synthesized
R2
HN
R1
R1
CO2Me
NH2
R1
R1
CO2Me
NH
R1
R1
O
a
b
NH
7a: R1 = H
b: R1 = MeO
3a-f
8a-b
N
N
Scheme 1. Reagents: (a) 4-pyridinecarboxaldehyde, NaCNBH3, MeOH, 8a: 69%, 8b: 79%; (b) RNH2, AlMe3, toluene, 30–71%.

H. Nakamura et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5127–5131
5129
Table 1. Yields and VEGFR1/2 kinase inhibitions of the anthranilic amides 3a–f, salicylic amides 4a–h, anthranilic amidines 5a–f, and salicylic
amidines 6a–f
R2
R2
R2
R2
O
HN
HN
O
HN
HN
O
R1
R1
R1
R1
R1
R1
R1
R1
NH
NH
O
NH
NH
3
4
5
N
6
N
N
N
Compound
R1
R2
Yield (%)
% of inhibitiona
KDRc
Flt-1b
—
3a
3b
3c
3d
3e
3f
H
71
30
41
59
30
53
—
30
—
—
—
—
N
N
H
12
—
—
—
—
N
N
H
N
N
MeO
MeO
MeO
N
N
N
N
4a
4b
4c
4d
4e
H
2-CF3C6H4
2-ClC6H4
2-CF3C6H4
2-ClC6H4
4-ClC6H4
35
20
35
20
20
50
—
—
—
3
60
24
—
—
55
H
MeO
MeO
MeO
N
4f
MeO
MeO
MeO
33
—
—
—
—
—
24
N
N
4g
4h
33
N
N
Trace
5a
5b
H
2-CF3C6H4
4-ClC6H4
2-ClC6H4
2-CF3C6H4
4-ClC6H4
2-ClC6H4
2-CF3C6H4
4-ClC6H4
2-ClC6H4
2-CF3C6H4
4-ClC6H4
2-ClC6H4
70
31
49
80
57
28
48
47
54
35
19
31
—
—
—
—
—
—
—
—
—
—
—
—
34
44
H
7
—
5c
5d
H
MeO
MeO
MeO
H
—
—
5e
5f
6a
—
—
6b
6c
H
—
—
H
6d
6e
MeO
MeO
MeO
—
—
6f
—
95 (31 nM)
AAL993d
a The inhibitory activity of compounds toward tyrosine kinases was determined by ELISA. Kinase assay was performed at a 10 lM concentration of
compounds.
b Catalytic domain of VEGFR1 protein tyrosine kinase.
c Catalytic domain of VEGFR2 protein tyrosine kinase.
d Kinase assay was performed at a 1 lM concentration of AAL993. IC50 value is indicated in parentheses.

5130
H. Nakamura et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5127–5131
O
O
R1
R1
CO2H
OMe
R1
R1
R2
R1
R1
R2
N
N
a, b
c
H
H
O
OH
9a: R1 = H
10a-f
4a-f
b: R1 = MeO
N
Scheme 2. Reagents: (a) i—SOCl2; ii—R2NH2, DMAP, pyridine, 50–99%; (b) BCl3, CH2Cl2, 23–78%; (c) i—KOH, H2O; ii—4-chloromethylpyridine,
20–35%.
O
MeO
MeO
CO2H
OMe
MeO
MeO
CO2Me
OH
MeO
MeO
R2
N
a
b,c
H
O
9b
4g-h
11
N
Scheme 3. Reagents and conditions: (a) i.—BCl3, CH2Cl2; ii—H2SO4, MeOH, reflux, 79%; (b) i—KOH, H2O; ii—4-chloromethylpyridine, 61%;
(c) RNH2, AlMe3, toluene, 4g: 33%, 4h: <5%.
R2
HN
R1
R1
R1
R1
R1
R1
CN
NH
CN
a
b
NH
NH
NH2
ALA881
GLU885
CYS1045
12a: R1 = H
5a-f
N
13a-b
N
b: R1 = MeO
Scheme 4. Reagents and conditions: (a) 4-pyridinecarboxaldehyde,
AcOH, MeOH, then NaCNBH3, rt, 12a: 41%, 12b: 92%; (b) RNH2,
AlMe3, toluene, 70 °C, 28–80%.
CYS919
R2
HN
Asp1046
R1
R1
R1
R1
R1
R1
CN
O
CN
OH
a
b
NH
O
14a: R1 = H
N
15a-b
6a-f
N
Figure 3. Docking modes of 4a (green) and 4e (gray) overlaid with the
binding conformation of AAL993 (orange) into the active site of
VEGFR2 kinase domain. Docking model was calculated by the DS
modeling 1.2 based on the X-ray analysis data of VEGFR2 tyrosine
kinase with aminopyrimidine 28 (PDB code: 1YWN). The yellow cloud
shows a ligand-binding site calculated by the site finding module.
b: R1 = MeO
Scheme 5. Reagents and conditions: (a) i—KOH, MeOH; ii—4-
chloromethylpyridine, Et3N, DMF, 140 °C, 14a: 41%, 14b: 46%; (b)
RNH2, AlMe3, toluene, 70 °C, 47–54%.
Figure. 3. The 4,5-dimethoxysalicylic group of 4e occu-
pied the binding site of CF3C6H4 groups of 4a and
AAL993, and the 4-ClC6H4 group was located near
Ala879. The ligscore2 of 4e was calculated as 4.78.
In order to better understand the structure-activity rela-
tionship between the VEGFR2 tyrosine kinase inhibi-
tion and the possible binding modes of the salicylic
amides 4a and 4e, we performed molecular docking
experiments of AAL993, 4a and 4e with the ligand-bind-
ing site of VEGFR2 kinase (PDB code 1YWN). Accord-
ing to our docking simulation as shown in Figure 3, 4a
would interact with VEGFR2 kinase domain through
hydrogen bonds between the pyridine–N and the back-
bone-NH of Cys919 in the hinge region of the kinase,
the salicylamide–C@O and the backbone–NH of
Asp1046 of the DFG-loop, and the salicylamide–NH
with the side-chain carboxylate of Glu885 of helix–C,
as indicated by the green stick; the ligscore215 was calcu-
lated as 5.33, which is smaller than that of AAL993 (the
ligscore2 = 5.86). The salicylic amide 4e showed a differ-
ent binding mode as indicated by gray bold stick in
In conclusion, we succeeded in the synthesis of a series
of anthranilic amides, salicylic amides, anthranilic ami-
dines, and salicylic amidines. Among the compounds
synthesized, the salicylic amide 4a possessed inhibitory
activities of both Flt-1 and KDR, whereas, the salicylic
amide 4e and the anthranilic amidine 5a were found as
selective inhibitors of VEGFR-2 tyrosine kinases. Two
methoxy groups substituted on the salicylic ring of 4e
enhanced the selectivity of VEGFR2 kinase inhibition
in comparison with 4a. We believe that the current find-
ing would be important for designs of tyrosine kinase-
targeted new therapeutic agents.

H. Nakamura et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5127–5131
5131
Sechler, J.; Fuentes-Pesquera, A.; Johnson, D.; Middle-
ton, S. A.; Jolliffe, L.; Chen, X. J. Med. Chem. 2005, 48,
1886.
Acknowledgments
This work was supported by the Ministry of Education,
Science, Sports, Culture and Technology, Grant-in-Aid
for the Screening Committee of New Anticancer Agents
on Priority Area ‘Cancer’ for the protein kinase assay
and for Scientific Research (A) for young scientists
(No. 16685017) from Japan.
8. Miyazaki, Y.; Matsunaga, S.; Tang, J.; Maeda, Y.;
Nakano, M.; Philippe, R. J.; Shibahara, M.; Liu, W.;
Sato, H.; Wang, L.; Nolte, R. T. Bioorg. Med.Chem. Lett.
2005, 15, 2203.
9. (a) Manley, P. W.; Bold, G.; Fendrich, G.; Furet, P.;
Mestan, J.; Meyer, T.; Meyhack, B.; Stark, W.; Strauss,
A.; Wood, J. Cell. Mol. Biol. Lett. 2003, 8, 532; (b)
Manley, P. W.; Bold, G.; Bruggen, J.; Fendrich, G.; Furet,
¨
References and notes
P.; Mestan, J.; Schnell, C.; Stolz, B.; Meyer, T.; Meyhack,
B.; Stark, W.; Strauss, A.; Wood, J. Biochim. Biophys.
Acta 2004, 1697, 17.
1. Risau, W. Nature 1997, 386, 671.
2. (a) Manley, P. W.; Bold, G.; Bruggen, J.; Frendrich, G.;
¨
10. (a) Asano, T.; Yoshikawa, T.; Nakamura, H.; Uehara,
Y.; Yamamoto, Y. Bioorg. Med. Chem. Lett. 2004, 14,
2299; (b) Asano, T.; Yoshikawa, T.; Usui, T.; Yamam-
oto, H.; Yamamoto, Y.; Uehara, Y.; Nakamura, H.
Bioorg. Med. Chem. 2004, 12, 3529; (c) Asano, T.;
Nakamura, H.; Uehara, Y.; Yamamoto, Y. ChemBio-
Chem 2004, 5, 483.
11. Dean, F. M.; Goodchild, J.; Houghton, J. A.; Martin, R.
B.; Parton, B.; Price, A. W.; Somvichien, N. Tetrahedron
Lett. 1966, 7, 4153.
Furet, P.; Mestan, J.; Schnell, C.; Stolz, B.; Meyer, T.;
Meyhack, B.; Stark, J.; Strauss, A.; Wood, J. Biochim.
Biophys. Acta 2004, 16975, 17; (b) Lawrence, D. S.; Niu, J.
Pharmacol. Ther. 1998, 77, 81.
3. (a) Abdollahi, A.; Lipson, K. E.; Han, X.; Krempien, R.;
Trinh, T.; Weber, K. J.; Hahnfeldt, P.; Hlatky, L.; Debus,
J.; Howlett, A. R.; Huber, P. E. Cancer Res. 2003, 63,
3755; (b) Sun, L.; Tran, N.; Liang, C.; Hubbard, S.; Tang,
F.; Lipson, K.; Schreck, R.; Zhou, Y.; McMahon, G.;
Tang, C. J. Med. Chem. 2000, 43, 2655.
12. Ibrahim, A. Y.; Behbehani, H.; Ibrahim, R. M.; Malhas,
R. Tetrahedron 2003, 59, 7273.
13. Iwata, K. K.; Jani, J. P.; Provoncha, K.; Kath, J. C.; Liu,
Z.; Moyer, J. D. Cancer Res. 2003, 63, 4450.
4. (a) Hennequin, L. F.; Stokes, E. S. E.; Thomas, A. P.;
´
Johnstone, C.; Ple, P. A.; Ogilvie, D. J.; Dukes, M.;
Wedge, S. R.; Kendrew, J.; Curwen, J. O. J. Med. Chem.
2002, 45, 1300; (b) Hennequin, L. F.; Thomas, A. P.;
14. EIA/RIA stripwellTM plates (Corning) were coated by
incubation overnight at 4 °C with 100 ll/well of 50 lg/ml
poly(Glu:Tyr,4:1) peptide (Sigma) in PBS. The kinase
reaction was performed in the plates by addition of
50 ll of kinase buffer (50 mM HEPES, 125 mM NaCl,
10 mM MgCl2, pH 7.4) containing 100 lM of ATP,
10 ng of recombinant Flt-1 or KDR (Invitrogen, cata-
lytic domain of VEGFR1 and VEGFR2, respectively),
and a compound (10 lM). After 20 min, the plates were
washed three times with wash buffer (0.1% Tween 20 in
PBS) and incubated for 20 min with 50 ll/well of 0.2 lg/
ml HRP conjugated anti-phosphotyrosine antibody
(Santa Cruz). After two washes, the plates were devel-
oped by addition of 50 ll/well Tetramethylbenzidine
(Sigma) and stopped by addition of 50 ll/well of 2 N
H2SO4. The absorbance at 450 nm was measured by a
96-well plate reader (Tecan).
´
Johnstone, C.; Stokes, E. S. E.; Ple, P. A.; Lohmann, J.-J.
M.; Ogilvie, D. J.; Dukes, M.; Wedge, S. R.; Kendrew, J.;
Lambert van der Brempt, C. J. Med. Chem. 1999, 42,
5369.
5. (a) Drevs, J.; Muller-Driver, R.; Wittig, C.; Fuxius, S.;
Esser, N.; Hugenschmidt, H.; Konerding, M. A.; Alleg-
rini, P. R.; Wood, J.; Hennig, J.; Unger, C.; Marme, D.
Cancer Res. 2002, 62, 4015; (b) Bold, G.; Altman, K.-H.;
Frei, J.; Lang, M.; Manley, P. W.; Traxler, P.; Wietfeld,
B.; Bruggen, J.; Buchdunger, E.; Cozens, R.; Ferrari, S.;
Furet, P.; Hofmann, F.; Martiny-Baron, G.; Mestan, J.;
Rosel, J.; Sills, M.; Stover, D.; Acemoglu, F.; Boss, E.;
Emmenegger, R.; Lasser, L.; Masso, E.; Roth, R.;
Schlachter, C.; Vetterli, W.; Wyss, D.; Wood, J. M.
J. Med. Chem. 2000, 43, 2310.
6. Manley, P. W.; Furet, P.; Bold, G.; Bruggen, J.; Mestan,
J.; Meyer, T.; Schnell, C. R.; Wood, J.; Haberey, M.;
Huth, A.; Kruger, M.; Menrad, A.; Ottow, E.; Seidel-
mann, D.; Siemeister, G.; Thierauch, K-H. J. Med. Chem.
2002, 45, 5687.
15. LigScore2 is a fast, simple scoring function for predicting
protein–ligand binding affinities. The functional form of
LigScore is simple and contains the CFF force field using
grid-based energy calculations with the improved interpo-
lation method; see: Wang, R.; Lu, Y.; Wang, S. J. Med.
Chem. 2003, 46, 2287.
7. Kou, G.-H.; Wang, A.; Emauel, S.; DeAngelis, A.; Zhang,
R.; Connolly, P. J.; Murray, W. V.; Gruninger, R. H.;