Discovery of [7-(2,6-dichlorophenyl)-5-methylbenzo [1,2,4]triazin-3-yl]-[4-(2-pyrrolidin-1-ylethoxy)phenyl]amine-a potent, orally active Src kinase inhibitor with anti-tumor activity in preclinical assays

Article abstract of DOI:10.1016/j.bmcl.2006.11.006

We describe the identification of [7-(2,6-dichlorophenyl)-5-methylbenzo [1,2,4]triazin-3-yl]-[4-(2-pyrrolidin-1-ylethoxy)phenyl]amine (3), a potent, orally active Src inhibitor with desirable PK properties, demonstrated activity in human tumor cell lines

Full text of DOI:10.1016/j.bmcl.2006.11.006

Bioorganic & Medicinal Chemistry Letters 17 (2007) 602–608
Discovery of [7-(2,6-dichlorophenyl)-5-methylbenzo
[1,2,4]triazin-3-yl]-[4-(2-pyrrolidin-1-ylethoxy)phenyl]-
amine—a potent, orally active Src kinase inhibitor
with anti-tumor activity in preclinical assays
Glenn Noronha,a,* Kathy Barrett,a Antonio Boccia,a Tessa Brodhag,a Jianguo Cao,a
Chun P. Chow,a Elena Dneprovskaia,a John Doukas,a Richard Fine,b Xianchang Gong,a
Colleen Gritzen,a Hong Gu,c Ehab Hanna,a John D. Hood,a Steven Hu,a Xinshan Kang,b
Jann Key,a Boris Klebansky,b Ahmed Kousba,a Ge Li,c Dan Lohse,a Chi Ching Mak,a
Andrew McPherson,a Moorthy S. S. Palanki,a Ved P. Pathak,a Joel Renick,a Feng Shi,c
Richard Soll,a Ute Splittgerber,a Silva Stoughton,a Suhan Tang,c Shiyin Yee,a
Binqi Zeng,a Ningning Zhaoa and Hong Zhua
aTargeGen, Inc., 9380 Judicial Drive, San Diego, CA 92121, USA
bBioPredict, Inc., 660 Kinderkamack Road, Oradell, NJ 07649, USA
cWuXi PharmaTech Ltd, Shanghai, PR China
Received 26 September 2006; revised 1 November 2006; accepted 2 November 2006
Available online 7 November 2006
Abstract—We describe the identification of [7-(2,6-dichlorophenyl)-5-methylbenzo [1,2,4]triazin-3-yl]-[4-(2-pyrrolidin-1-yleth-
oxy)phenyl]amine (3), a potent, orally active Src inhibitor with desirable PK properties, demonstrated activity in human tumor cell
lines and in animal models of tumor growth.
ꢀ 2006 Elsevier Ltd. All rights reserved.
The Src-family of tyrosine kinases, of which Src is the
best known, comprises the highly homologous proteins
Src, Yes, Fyn, Lyn, Hck, Blk, Brk, Fgr, Frk, Srm,
Lck, and Yrk.1 Gene knockout studies of specific Src-
family members have elucidated the roles played in
bones, neural development, and myeloid and T cell
development and function.2 Src has been shown to play
a role in myocardial infarction,3 stroke,4 osteoporosis,5
neurodegeneration,6 and is involved in metastases and
tumor progression, particularly those of breast, meta-
static colorectal,7 ovarian,8 and pancreatic cancers.9–11
Despite evidence of the dysregulation of Src in several
cancers, there is no known Src inhibitor approved for
solid tumors. Dasatinib, a potent dual inhibitor of Src
and Abl, was recently approved for the oral treatment
of chronic myelogenous leukemia (CML)12 and Phila-
delphia chromosome-positive acute lymphoblastic leu-
kemia (ALL). Currently Dasatinib is in clinical trials
for the treatment of solid tumors and gastrointestinal
stromal tumors (GIST). On account of the pivotal role
played by Src in several oncology and non-oncology dis-
ease states, inhibitors of Src have broad potential utility.
Several series of Src inhibitors have been reported.13–20
We have recently disclosed a novel series of benzotri-
azine-based compounds as inhibitors of Src.21 Herein,
we describe our continuing efforts in optimizing this
benzotriazine series of Src inhibitors, and the identifica-
tion of an orally active compound with desirable PK
properties, activity against selected human tumor cell
lines, and demonstrated efficacy in mouse models of
tumor growth.
Keywords: Src inhibitor; Cancer; Kinase inhibitor; Benzotriazines.
*
Preliminary SAR studies enabled the identification of
key structural requirements of the benzotriazine series
Corresponding author. Tel.: +1 858 678 0760; fax: +1 858 678 0762;
e-mail: noronha@targegen.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.006

G. Noronha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 602–608
603
that lead to substitutions at the 3-, 5- or 6-, and 7-posi-
tions critical for obtaining low nanomolar biochemical
potencies.21 A homology model was constructed for a
fully activated Src structure that served to guide com-
pound design and optimization.22 Figure 1 depicts a
minimized binding mode of a benzotriazine inhibitor
(1) in the ATP pocket of the activated Src model. We
have subdivided this ATP-binding pocket of Src into
four distinct regions, labeled P1 through P4. The bind-
ing interactions of 1 at the ATP pocket are summarized
as follows. The 7-substituted phenyl ring binds in the
hydrophobic back pocket (the P4 region) positioning
this phenyl ring perpendicular to the benzotriazine core.
Our earlier studies21 showed that a limited set of small
hydrophobic groups may be substituted on this phenyl
ring at very specific positions since the P4 pocket pos-
sesses a limited tolerance for larger groups. The 3-amin-
obenzotriazine moiety forms both donor and acceptor
hydrogen bonds to Met-341 in the hinge region (termed
P3), thus sharing a common site with the hinge binding
purine ring of ATP. The 3-amino group from the benzo-
triazine core is attached to an aryl linker in a region of
the ATP pocket termed P2, which falls in the hydropho-
bic channel leading out toward a solvent-accessible area.
The spacer with a solubilizing group binds in the P1 re-
gion of the pocket, an area accessible to solvent. Benzo-
triazine inhibitors containing a basic nitrogen in the P1
region provide enhanced potency through protonation
and interaction with Asp-348.
(1), and in the second compound through a sulfonamide
linker (17). Both linkers place the 1-ethyl pyrrolidine
group in the P1 region where the basic nitrogen of the
pyrrolidine interacts with Asp-348.
Initial SAR efforts were focused on compounds that
have their ethylpyrrolidine moiety in the P1 region, at-
tached to an aryl group (positioned in the P2 region)
via an oxygen linker (ether series, Table 1) and exploring
various substituents at the 7-position (P4 region) of the
benzotriazine ring.23 The removal of the dimethyl
groups on the phenyl ring of 1 results in compound 2
with a 50-fold decrease in potency. The replacement of
the dimethyl groups on the 7-phenyl ring in 1 with
dichloro resulted in compound 3 with comparable
potency. The monochloro analog 4 was three times less
potent than the dichloro analog 3. Compounds with
either a 2,6-difluorophenyl substituent at the 7-position
(5) or a 2-chloro-6-methoxy phenyl at the 7-position
(6) were 30 times and 70 times less potent than 1. It ap-
pears that fluorine atoms are too small and not as effec-
tive as chloro or methyl groups to orinent the phenyl
group perpendicular to the benzotriazine ring, a requi-
site for low nanomolar potency. A methoxy group at
the 6-position on the 7-phenyl ring orients the phenyl
group perpendicular to the benzotriazine ring, however,
the methoxy group may be sterically hindering the bind-
ing of the 7-phenyl group within this P4 site (6). The
binding pocket in the P4 region is tolerant toward small
substituents on the 7-phenyl ring, as in the case of
methyl groups or chloro groups. The introduction of
heteroaromatics in place of the 7-phenyl ring results in
decreases in potency as seen with compounds 9–12.
The removal of the 5-methyl group on the benzotriazine
ring results in a minor loss of potency as seen in com-
pounds 13 and 14. Both compounds with 5-methyl (1)
and 6-methyl (16) groups on the benzotriazine core
share similar potencies.
Motivated by a preliminary optimization of the benzo-
triazine-based Src inhibitors,21 we focused our efforts
on two compounds. Both possess the 1-ethylpyrrolidinyl
moiety, which binds in the P1 region. In the first com-
pound, the 1-ethylpyrrolidinyl group is attached to the
3-aminophenyl ring (P2 region) through an ether linker
Table 1. In vitro evaluation of compounds with R1 and R2 modifi-
cations on the 3-aminobenzo-[1,2,4]-triazine
R1
N
N
O
N
N
6
R2
N
5
2.13
2.06
H
Compound
R1
R2
Src (IC50 lM)
1
2
2,6-Me2 Ph
Ph
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
H
0.007
0.341
0.011
0.035
0.208
0.505
0.778
1.20
3
2,6-Cl2 Ph
2-Cl Ph
2,6-F2 Ph
P4
P3
P2
P1
4
5
1
6
2-Cl,6-OMe Ph
2-Cl,5-OMe Ph
3-Cl,6-OMe Ph
4-(2-F, 5-Me pyridyl)
4-(3,5-Me2 isoxazole)
4-Pyrimidyl
7
N
O
N
3
N
7
8
N
N
5
9
0.204
0.207
1.5
H
10
11
12
13
14
15
16
Hydrophobic Hinge binding Aryl
region region linker
Solvent
accessible
2-Pyridyl
2,6-Me2 Ph
2,6-Cl2 Ph
1.8
0.015
0.030
>10
region region
H
H
2,6-Me2 Ph
H
6-Me
Figure 1. A homology model of the ATP pocket of Src, showing the
binding of 1.
0.012

604
G. Noronha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 602–608
Table 2. In vitro evaluation of compounds with R1 and R2 modifi-
cations on the 3-aminobenzo-[1,2,4]-triazine
compared to 17. Moreover, when the ortho chloro sub-
stituent in 21 was removed as in compound 22, we saw
a 5-fold decrease in potency compared to 21 and a 30-
fold drop in potency compared to 17. These observa-
tions are consistent with the SAR from the ether series.
It appears that the P4 region accommodates phenyl
groups with small substituents at the 2- and the 6-posi-
tions. The position of the methyl group (5- or 6-) on the
benzotriazine ring did not change the potency as seen
with compounds 26 and 17. The SAR of the 6-methyl
series is comparable to that seen in the 5-methyl series.
Removal of the bis ortho substitutions on the phenyl
ring results in compound 27 with a 30-fold loss of poten-
cy compared to 17 and 26. Removal of the 7-phenyl
moeity (28) or substitution with a bromine (29) results
in a complete loss of potency. The 2-methylphenyl ana-
log 30 showed comparable potency to 17, while the 2-tri-
fluoromethyl analog 31 suffered ca. 6-fold loss of
potency compared to 30 and 17. Similarly, the 3-cya-
no-6-methyl analog 32 shares similar potency to the cor-
responding 5-methyl analog, 20. Similar to the 5-methyl
benzotriazine analogs, ortho substitutions on the
7-phenyl ring in the 6-methyl benzotriazine compounds
play an important role in maintaining potency.
O
O
R1
N
N
S
N
N
N
H
6
R2
N
5
H
Compound
R1
R2
Src (IC50 lM)
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
2,6-Me2 Ph
2,6-Cl2 Ph
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
5-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
6-Me
0.009
0.014
0.013
0.025
0.051
0.282
0.353
0.631
2.90
2-Cl, 6-OH Ph
3-CN, 6-Me Ph
3-NH2, 6-Cl Ph
3-NH2
3-CH2NH2, 6-Me Ph
3-NHMe Ph
3-NHSO2Me Ph
2,6-Me2 Ph
Ph
0.011
0.315
>10
H
Br
>10
0.01
2-MePh
2-CF3Ph
0.067
0.012
0.020
0.092
0.072
3-CN, 6-Me Ph
4-F, 2-Me Ph
2-NH2
2-CH2OH Ph
It is evident from the above SAR, that the 7-(2,6-di-
methyl)phenyl substituted analogs are some of the most
potent inhibitors. Therefore, we continued our lead opti-
mization efforts while keeping this 2,6-dimethylphenyl
moiety fixed in the P4 region, and exploring various
water-soluble groups in the P1 region of the molecule
(Table 3).23 A 2-diethylaminoethoxy group either at
the 3-position or at 4-position of the phenyl group in
P2 region resulted in 36 and 37, with comparable poten-
cy to 1. A 2-morpholin-4-ylethoxy group at the 3-posi-
tion also resulted in a compound (38) with comparable
potency to 1. Extending the pyrrolidinyl group in 1 by
a methylene unit resulted in compound 39 with compa-
rable potency. Similarly moving the pyrrolidinylethoxy
side chain from the 4- to the 3-position on the phenyl
In parallel, we focused our efforts on optimizing com-
pounds in the sulfonamide series by substituting various
groups at the 7-position of benzotriazine ring in P4 re-
gion (Table 2). The replacement of the ether linkage in
1 with a sulfonamide group resulted in 17 with compa-
rable potency to 1. The substitution of dichloro groups
on the phenyl ring in place of the dimethyl groups result-
ed in 18 with comparable potency to 17 and 3. The
replacement of one of the chloro groups with a hydroxy
group resulted in 19 without any change in the potency.
When one of the ortho substituents was moved to meta
position and substituted with a cyano (20) or an amino
group (21), this resulted in a 3- to 5-fold loss in potency
Table 3. In vitro evaluation of compounds with R4 modifications on the 3-aminobenzo-[1,2,4]-triazine
3
N
4
N
X
N
N
R3
H
Compound
X
R3
Src (IC50 lM)
36
37
38
39
40
41
42
43
44
45
46
47
48
49
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
N
3-(2-Diethylaminoethoxy)
4-(2-Diethylaminoethoxy)
0.010
0.010
0.033
0.013
0.0064
0.021
0.040
0.036
0.039
0.015
0.013
0.016
0.018
0.0064
3-(2-Morpholin-4-ylethoxy)
4-(3-Pyrrolidin-1-yl-propoxy)
3-(2-Pyrrolidin-1-yl-ethoxy)
4-[2-(4-Methyl-piperazin-1-yl)-ethoxy]
3-[2-(4-Methyl-piperazin-1-yl)-ethoxy]
4-[2-(4-Methyl-piperazin-1-yl)-propoxy]
3-[2-(4-Methyl-piperazin-1-yl)-propoxy]
4-(40-Methyl-piperazin-1-yl)-amido
4-(Piperazin-1-yl)-amido
4-(N-(2-Pyrrolidin-1-yl-ethyl)-benzamide)
3-(N-(2-Pyrrolidin-1-yl-ethyl)-benzamide)
3-(2-Pyrrolidin-1-yl-ethoxy)

G. Noronha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 602–608
605
group in the P2 region resulted in compound 40 with
comparable potency. The 4-methylpiperazinyl group
with an ethoxy linker at the 4-position (41) or at the
3-position (42) resulted in compounds with comparable
potencies. The linker in compounds 41 and 42 was
extended by one more methylene group to give com-
pounds 43 and 44, respectively, with no change in enzy-
matic activity. The oxygen linkers in compounds 1 and
40 were substituted with amide linkers resulting in com-
pounds 47 and 48, once again with comparable activity.
Substitution of the phenyl group in the P2 region with a
pyridyl ring resulted in 49 with equally good potency. It
appears that the P1 region is tolerant toward a wide
number of water-soluble groups.
The benzotriazines studied were synthesized as shown in
Scheme 1. Appropriately substituted 3-bromo-2-nitro-
anilines (54) were reacted with cyanamide under acidic
conditions to give intermediate guanidines, which were
cyclized with sodium hydroxide to give 1-oxobenzotria-
zines (55).24,25 These N-oxides (55) were reduced using
Raney nickel to give 3-amino-7-bromobenzotriazines
in good yield. Various aryl-substituted benzotriazines
(56) were prepared by treating 3-amino-7-bromobenzo-
triazines with aryl boronic acids under Suzuki coupling
conditions.26 The final compounds (1–49) were prepared
from 56 in good yield by using palladium- and xant-
phos-catalyzed amination.27 The synthesis of 50–53
started with 57, which was prepared from 54, as de-
scribed above. The nitration of 57 resulted in 58 with
exclusive nitration at 5-position in good yield.
Oxobenzotriazines 58 were subjected to Suzuki and pal-
ladium-catalyzed amination couplings as described
above followed by the reduction of the nitro group to
amine (60) using 10% Pd on carbon under hydrogen
atmosphere. The amine group in 60 was functionalized
with appropriate alkyl halides or acid chlorides to give
50–53 in good yield.
Modeling studies of the benzotriazine inhibitors suggest
that the 5- and 6-positions of the benzotriazine ring in
the P3 region are oriented toward the sugar binding
pocket and away from the hinge region, and potentially
this portion of the inhibitors might be tuned to take
advantage of potential binding interactions and to pro-
vide for property diversification. A limited number of
hydrophilic groups were substituted in place of the
5-methyl group in order to explore this space in the
enzyme. Substitution of the 5-methyl group with an ami-
no (50), an, methylamino (51), an acetamido (52), or a
methylsulfonamido (53) moiety all resulted in com-
pounds with comparable potency to 1 (Table 4).
Several of these potent compounds shown in Tables 1–3
were tested in a murine CT-26 carcinoma cell assay for
further evaluation.28 Based on the potency in the cell
and the biochemical assays, eight compounds (with
IC50 less than 10 lM in the cell assay) were selected
for additional profiling. The cell data for these eight
compounds are collected beneath the graph in Figure 2.
Table 4. In vitro evaluation of compounds with R2 modifications on
the 3-aminobenzo-[1,2,4]-triazine
The selected compounds were then evaluated in a pul-
monary metastases model.29,30 Control mice treated
with vehicle showed extensive tumor burden in the lung
(Fig. 2). Compounds were administered to mice (50 mg/
kg, i.p., q3d) and all the compounds showed reduction
in tumor volume compared to vehicle control.31 The
greatest reduction in tumor volume (60%) compared to
vehicle control was seen with compound 3. Compounds
1, 45, and 49 showed equal reduction in tumor burden.
N
N
O
N
N
N
H
R2
Compound
R2
Src (IC50 lM)
50
51
52
53
NH2
NHMe
0.010
0.010
0.014
0.021
NHCOCH3
NHSO2Me
Based on the above initial screening, compounds 1 and 3
were selected for protein binding assays and pharmaco-
Scheme 1. Reagents and conditions: (a) NH2CN, 37% aq HCl, 110 ꢁC, 1.5 h; (b) 30% aq NaOH, 110 ꢁC, 0.5 h, 65% over two steps; (c) 10% Raney
Ni, H2, EtOH, 22 ꢁC, 4 h, 70–80%; (d) R1B(OH)2, Pd(PPh3)4, K2CO3, DME/EtOH/H2O (4:1:1), D, 3 h, 80%; (e) R3Ar or HetArBr, Pd2(dba)3,
xantphos, Cs2CO3, dioxane, D, 12 h, 50–70%; (f) H2SO4/HNO3 (1:1), 22 ꢁC, 16 h, 95%; (g) H2, 10% Pd/C, EtOH, 22 ꢁC, 12 h, 70–80%; (h) R2Cl,
CH2Cl2, 22 ꢁC, 1 h, 70–80%.

606
G. Noronha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 602–608
Table 6. In vitro evaluation of three compounds in various CYP
isozymes
800
600
400
200
Compound
CYP isozymes (IC50, lM)
3A4
1A2
2C9
2C19
2D6
1
3
42
62
>100
>100
>100
>100
12
12
25
33
0
Compound 3 was evaluated in an A549 human NSCLC
xenograft model (Fig. 3),33 along with Tarceva34 as posi-
tive control.35 At 25 mg/kg, i.p., qod, compound 3
showed potency comparable to Tarceva at 80 mg/kg,
po, qod. Moreover, at the higher dose of 40 mg/kg,
compound 3 showed 86% reduction in tumor burden
compared to the vehicle control. In order to establish a
link between the molecular target of 3 and its pharmaco-
dynamic effects, the impact of 3 on Src kinase activities
was examined in tumor tissue in vivo.36 The Tyr-861
residue of Focal Adhesion Kinase (Fak) is a specific sub-
strate for Src.37–39 Compound 3 inhibited phosphoryla-
tion of Fak-Y861 in these tumors in a dose-dependent
manner, with full inhibition at 40 mg/kg (Fig. 3), suggest-
ing that compound 3 was inhibiting Src in vivo.40 This
in vivo inhibition of the phosphorylation of Fak-Y861
in a dose-dependent manner suggests that Fak-Y861
has potential utility as a biomarker of compound activity
in clinical biopsy samples.
vehicle 47 36 41
1
45 49
3
Compd
(Dose, mg/kg)
Cell data, IC50, μM
(50) (50) (50) (50) (50) (50) (50)
2.4 2.5 6.6 2.3 6.7 2.4 0.75
Figure 2. Tumor burden model with murine CT-26 carcinoma cells.
Compounds were dosed ip, q3d. The IC50 values for these compounds
in murine CT-26 carcinoma cell line are shown.
kinetic (PK) studies (Table 5). Both compounds showed
high protein binding to human plasma.32 We examined
the PK properties of compounds 1 and 3 in mouse,
rat, and dog. Both compounds showed Cmax between
300 and 1000 ng/mL (ca. 0.6–2 lM) in all three species
with the exception of compound 1 in the dog where
the Cmax was 5240 ng/mL (>10 lM). Both compounds
have favorable bioavailability (74–86%) in mouse plas-
ma after oral dosing. Compounds 1 and 3 showed com-
parable PK data in rat and differ significantly in dog.
These compounds have very long T1/2 in dog, followed
by mouse and moderate T1/2 in rat. The clearance and
volume of distribution of 3 in dog were much higher
than those shown by 1. Both compounds were evaluated
for their Cyp inhibition profile and found to be very
weak inhibitors of various Cyp isozymes (Table 6).
Based on the PK parameters and activity in the above
screening tumor model, compound 3 was selected for
further characterization.
Compound 3 was evaluated in a selected panel of ki-
nases (Table 7) and in general showed a low level of
selectivity against closely related Src family members,
other structurally related intracellular kinases, and a
moderate level of selectivity versus selected serine threo-
nine and certain receptor kinases. Compound 3 was also
evaluated in a select panel of cell-based assays for its
Table 5. In vitro evaluation in protein binding assay, and in vivo PK parameters (25 mg/kg, po)
Compound Protein binding Species Cmax (ng/mL) Tmax (h) T1/2 (h) AUC (h ng/mL) CLsys (mL/min/kg) Vdss (L/kg) F (%)
human plasma
1
1
1
3
3
3
>99.9%
99.8%
Mouse
Rat
Dog
790
419
6
9.5
5.6
19
8
19,070
9570
9680
18.8
9.2
11.4
7.2
86
20
68
74
35
42
6
5240
415
1.5
8
5.1
8.6
Mouse
Rat
Dog
15,300
8010
8680
12.7
12.7
14.5
16.1
9.4
348
679
8.8
4
4.6
13
18.7
Vehicle
A
B
3, 25 mg/kg, ip,qod
3, 40 mg/kg, ip, qod
Tarceva, 80 mg/kg, po, qd
800
Compound 3
Start dosing
600
400
200
Dose (mg/kg) Vehicle
5
25
40
0
0
5
10
15
Days
Figure 3. (A) Compound 3 inhibited the growth of A549 human NSCLC xenografts. The compound was administered ip starting on day 5.
(B) Proteins from the tumors were examined by immunoblotting using a polyclonal antibody directed against phospho-Focal Adhesion Kinase
(Fak)-Y861.

G. Noronha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 602–608
607
Table 7. In vitro evaluation of 3 in a selected panel of kinases
12. Kanerva, J.; Nwawka, O.; Hwang1, K.; Lee, F. Y.; Corey,
S. J. Blood 2004, 104, Abstract 4480.
Kinase
Ki (nM)
Kinase
Ki (nM)
13. Boschelli, D. H.; Wu, B.; Sosa, A. C. B.; Durutlic, H.; Ye,
F.; Raifeld, Y.; Golas, J. M.; Boschelli, F. J. Med. Chem.
2004, 47, 6666.
14. Boschelli, D. H.; Wu, B.; Sosa, A. C. B.; Durutlic, H.;
Chen, J. J.; Wang, Y.; Golas, J. M.; Lucas, J.; Boschelli, F.
J. Med. Chem. 2005, 48, 3891.
Src
28.1 2.1
12.9 1.9
29.4 3.4
25.9 1.2
19.4 3.5
VEGF-R2
FGF-R1
PDGFRb
EphB4
784 96
493 102
2090 670
63.8 5.2
407 109
Yes
Lck
Lyn
Abl
Ret
15. Boschelli, D. H.; Wu, B.; Sosa, A. C. B.; Chen, J. J.;
Golas, J. M.; Boschelli, F. Bioorg. Med. Chem. Lett. 2005,
15, 4681.
Table 8. In vitro evaluation of 3 in selected tumor cell lines
Cell line
Tumor type
IC50 lM
16. Manetti, F.; Locatelli, G. A.; Maga, G.; Schenone, S.;
Modugno, M.; Forli, S.; Corelli, F.; Botta, M. J. Med.
Chem. 2006, 49, 3278.
17. Carraro, F.; Naldini, A.; Pucci, A.; Locatelli, G. A.;
Maga, G.; Schenone, S.; Bruno, O.; Ranise, A.; Bondav-
alli, F.; Brullo, C.; Fossa, P.; Menozzi, G.; Mosti, L.;
Modugno, M.; Tintori, C.; Manetti, F.; Botta, M. J. Med.
Chem. 2006, 49, 1549.
18. Dalgarno, D.; Stehle, T.; Narula, S.; Schelling, P.; van
Schravendijk, M.; Adams, S.; Andrade, L.; Keats, J.;
Ram, M.; Jin, L.; Grossman, T.; MacNeil, I.; Metcalf, C.;
Shakespeare, W.; Wang, Y.; Keenan, T.; Sundaramoorthi,
R.; Bohacek, R.; Weigele, M.; Sawyer, T. Chem. Biol.
Drug Des. 2006, 67, 46.
19. Wang, Y.; Metcalf, C.; Shakespeare, W.; Sundaramoorthi,
R.; Keenan, T.; Bohacek, R.; Van Schravendijk, M.;
Violette, S.; Narula, S.; Dalgarno, D.; Haraldson, C.;
Keats, J.; Liou, S.; Mani, U.; Pradeepan, S.; Ram, M.;
Adams, S.; Weigele, M.; Sawyer, T. Bioorg. Med. Chem.
Lett. 2003, 13, 3067.
20. Honold, K.; Kaluza, K.; Masjost, B.; Schaefer, W.;
Scheiblich, S. Hoffmann-La Roche, WO2006066913, 2006.
21. Noronha, G.; Barrett, K.; Cao, J.; Dneprovskaia, E.; Fine,
R.; Gong, X.; Gritzen, C.; Hood, J.; Kang, X.; Klebansky,
B.; Lohse, D.; Mak, C. C.; McPherson, A.; Palanki, M. S.
S.; Pathak, V. P.; Renick, J.; Soll, R.; Splittgerber, U.;
Wrasidlo, W.; Zeng, B.; Zhao, N. Bioorg. Med. Chem.
Lett. 2006, 16, 5546.
22. The fully activated catalytic domain of human Src kinase
was built based on available crystal structures of activated
Lck. The catalytic domains of Src and Lck exhibit 67%
sequence identity over 273 residues without gaps in the
alignment with significantly higher homology exhibited in
the ATP-binding site. The model was built using the
interactive programs InsightII and Homology from Accel-
rys. Ligands were docked initially into the active site using
an automated modeling program BioDock (NIH SBIR
Grant 1R43GM071055) and subsequently using interac-
tive modeling in InsightII. Models with and without
ligands were subjected to energetic refinement including
solvent using the Accelerys program Discover. Visualiza-
tion was generated using the program BioInterpreter (NIH
SBIR Grant 5R44GM061465).
23. A portion of the material presented in this paper was
disclosed in two posters. (a) McPherson, A.; Barrett, K.;
Cao, J.; Gritzen, C.; Hood, J.; Mak, C. C.; Noronha, G.;
Pathak, V. P.; Renick, J.; Soll, R.; Splittgerber, U.; Zeng,
B. Abstracts of Papers, 231st National Meeting of the
American Chemical Society, Atlanta, GA, March 26–30,
2006; American Chemical Society: Washington, DC, 2006;
(b) Mak, C. C.; Barrett, K.; Cao, J.; Gritzen, C.; Hood, J.;
McPherson, A.: Noronha, G.; Pathak, V. P.; Renick, J.;
Soll, R.; Splittgerber, U.; Zeng, B. Abstracts of Papers,
231st National Meeting of the American Chemical Soci-
ety, Atlanta, GA, March 26–30, 2006; American Chemical
Society: Washington, DC, 2006; MEDI-065.
Molt-4
K562
Leukemia
Leukemia
Colon
0.250
0.215
0.750
2.50
CT-26
Colo-205
HT-29
Colon
Colon
>5.00
0.240
0.085
0.900
MDA-MB231
MCF-7
A549
Breast
Breast
NSCLC
selectivity (Table 8). The potency of compound 3 was
less than 1 lM in all the cell lines tested except Colo-
205 (IC50 = 2.5 lM) and HT-29 (IC50 > 5.0 lM).
In summary, this work describes efforts that led to an in-
creased understanding of the binding of these benzotri-
azines in the ATP pocket of Src, and resulted in
marked improvements in potency, and improved PK
properties. Further, lead optimization efforts unveiled
the activity of 3 against selected human tumor cell lines,
and demonstrated efficacy of 3 in mouse xenograft mod-
els driven by Src mediated pathways. Compound 3 has
potential use in the treatment of cancers of the colon,
liver, pancreas, breast, brain, and bladder, where Src
activity is dysregulated.41 Detailed studies on systemic
PK, metabolism, tolerability, and toxicity of compound
3 were completed and will be disclosed separately.
References and notes
1. Trevino, J. G.; Summy, J. M.; Gallick, G. E. Mini Rev.
Med. Chem. 2006, 6, 109.
2. Lowell, C. A.; Soriano, P. Genes Dev. 1996, 10, 1845.
3. Weis, S.; Shintani, S.; Weber, A.; Kirchmair, R.; Wood,
M.; Cravens, A.; McSharry, H.; Iwakura, A.; Yoon, Y.;
Himes, N.; Burstein, D.; Doukas, J.; Soll, R.; Losordo, D.;
Cheresh, D. J. Clin. Invest. 2004, 113, 885.
4. Paul, R.; Zhang, Z. G.; Eliceiri, B. P.; Jiang, Q.; Boccia, A.
D.; Zhang, R. L.; Chopp, M.; Cheresh, D. A. Nat. Med.
2001, 7, 222.
5. Susva, M.; Missbach, M.; Green, J. Trends Pharmacol.
Sci. 2000, 21, 489.
6. Yeatman, T. J. Nat. Rev. 2004, 4, 470.
7. Talamonti, M. S.; Roh, M. S.; Curley, S. A.; Gallick, G.
E. J. Clin. Invest. 1993, 91, 53.
8. Wiener, J. R.; Windham, T. C.; Estrella, V. C.; Parikh, N.
U.; Thall, P. F.; Deavers, M. T.; Bast, R. C.; Mills, G. B.;
Gallick, G. E. Gynecol. Oncol. 2003, 88, 73.
9. Ito, H.; Gardner-Thorpe, J.; Zinner, M. J.; Ashley, S. W.;
Whang, E. E. Surgery 2003, 134, 221.
10. Summay, J. M.; Gallick, G. E. Cancer Metastasis 2003, 22,
337.
11. Irby, R. B.; Yeatman, T. J. Oncogene 2000, 19, 5636.
24. Arndt, F. Ber. 1913, 46, 3522.
25. Mason, J. C.; Tennant, G. J. Chem. Soc. (B) 1970, 911.

608
G. Noronha et al. / Bioorg. Med. Chem. Lett. 17 (2007) 602–608
26. Kudo, N.; Perseghini, M.; Fu, G. C. Angew. Chem., Int.
Ed. 2006, 45, 1282.
27. Tundel, R. E.; Anderson, K. W.; Buchwald, S. L. J. Org.
Chem. 2006, 71, 430.
volumes were calculated using caliper measurements and
the ellipsoid formula. The experiment was ended when
vehicle-treated tumors showed signs of necrosis.
34. Ji, H.; Zhao, X.; Yuza, Y.; Shimamura, T.; Li, D.;
Protopopov, A.; Jung, B. L.; McNamara, K.; Xia, H.;
Glatt, K. A.; Thomas, R. K.; Sasaki, H.; Horner, J. W.;
Eck, M.; Mitchell, A.; Sun, Y.; Al-Hashem, R.; Bronson,
R. T.; Rabindran, S. K.; Discafani, C. M.; Maher, E.;
Shapiro, G. I.; Meyerson, M.; Wong, K. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 7817.
35. Currently Tarceva is recommended as a monotherapy for
the treatment of patients with locally advanced or meta-
static non-small cell lung cancer (NSCLC) after failure of
at least one prior chemotherapy regimen. Tarceva is an
approved agent for treating NSCLC. Therefore, it is
important to test the new compounds for treating NSCLC
against the approved and established therapeutic agent,
Tarceva.
28. Approximately 3000 CT-26 cells were added to 96-well
plates suspended in RPMI 1640 medium containing 10%
FBS, penicillin, streptomycin, and glutamine. Com-
pounds, dissolved in DMSO and ranging in concentration
from 5 to 10,000 nM, were added to a total of 1% of the
total assay volume. After these assays were incubated in a
10% carbon dioxide environment for 72 h at 37 C, XTT
was added and allowed to proceed for an additional 4 h.
Remaining cellular activity was assessed by reading the
absorbance at 492. After subtracting background absor-
bance at 650 nm, data analysis was done by using the
nonlinear graphing capabilities of GraphPad PRIZM,
Ver. 4.01.
29. Hood, J. D.; Bednarski, M.; Frausto, R.; Guccione, S.;
Reisfeld, R. A.; Xiang, R.; Cheresh, D. A. Science 2002,
296, 2404.
36. Colon tumor xenografts were formed by injecting 106
Colo-205 cells subcutaneously on the flank of nude mice
(nu/nu) (n = 5). By day 5, mean tumor size (as measured
using hand calipers) was 100–150 mm3 using the ellipsoid
formula. At this point animals were randomized so that all
groups had approximately the same mean tumor volume
and compound was administered at via oral gavage in a
Phosal 50 PG formulation.
37. Calalb, M.; Zhang, X.; Polte, T.; Hanks, S. Biochem.
Biophys. Res. Commun. 1996, 228, 662.
38. Lim, Y.; Han, I.; Jeon, J.; Park, H.; Bahk, Y.; Oh, E.
J. Biol. Chem. 2004, 279, 29060.
30. In order to establish pulmonary metastases model, murine
CT-26 carcinoma cells were injected intravenously into
BALB/c mice. This typically results in the formation of
lung metastases within four days. In this study, establish-
ment of pulmonary metastases was extended to 10 days
before treatment with the compounds to ensure that all
animals contained actively growing lung tumors.
31. The frequency and the amount of doses for this study were
chosen based on the preliminary PK studies on these
compounds (data not shown). These are not minimum
tolerated doses.
39. Eliceiri, B.; Paul, R.; Schwartzberg, P.; Hood, J. D.; Leng,
J.; Cheresh, D. Mol. Cell 1999, 4, 915.
32. Eriksson, M. A. L.; Gabrielsson, J.; Nilsson, L. B. J.
Pharm. Biomed. Anal. 2005, 38, 381.
40. Colo-205 tumors were subcutaneously implanted in mice
and allowed to grow to 100–150 mm3. Six hours after the
compound 3 was administered orally, tumors were resect-
ed from the animals for Western blot analysis using a
polyclonal antibody directed against Fak-Y861.
41. Tsygankov, A. Y.; Shore, S. K. Curr. Pharm. Des. 2004,
10, 1745.
33. A549 tumors were formed by implanting 1 · 106 cells
subcutaneously on the flanks of immunocompromised
(nu/nu) mice (n = 6). When the tumors reached 100 mm3,
compounds were administered ip in a solution of Solu-
tol:PEG:EtOH:Water = 10:10:10:70 formulation on day 5
using every other day (QOD) dosing schedule. Tumor