Inhibition of Src kinase activity by 4-anilino-5,10-dihydropyrimido[4,5-b]quinolines

Article abstract of DOI:10.1016/S0960-894X(03)00628-0

4-(2,4-Dichloro-5-methoxy)anilino-5,10-dihydropyrimido[4,5-b]quinolines are potent inhibitors of Src kinase and Src cellular activity while having no effect on Fyn cellular activity. The corresponding 4-(2,4-dichloro-5-methoxy)anilino-pyrimido[4,5-b]quino

Full text of DOI:10.1016/S0960-894X(03)00628-0

Bioorganic & Medicinal ChemistryLetters 13 (2003) 2977–2980
Inhibition of Src Kinase Activity by 4-Anilino-5,10-dihydro-
pyrimido[4,5-b]quinolines
Diane H. Boschelli,a,* Dennis Powell,a Jennifer M. Golasb and Frank Boschellib
aChemical and Screening Sciences, Wyeth Research, 401 N. Middletown Road, Pearl River, NY 10965, USA
bOncology, Wyeth Research, 401 N. Middletown Road, Pearl River, NY 10965, USA
Received 29 January2003; accepted 10 March 2003
Abstract—4-(2,4-Dichloro-5-methoxy)anilino-5,10-dihydropyrimido[4,5-b]quinolines are potent inhibitors of Src kinase and Src
cellular activity while having no effect on Fyn cellular activity. The corresponding 4-(2,4-dichloro-5-methoxy)anilino-pyrimido[4,5-b]
quinolines are much less effective Src inhibitors.
# 2003 Elsevier Ltd. All rights reserved.
Protein tyrosine kinases (TKs) catalyze the transfer of a
phosphate group from ATP to a specific tyrosine resi-
due on a protein. Since aberrant TK activityis asso-
ciated with a varietyof disease states, there is great
interest in investigating low molecular weight inhibitors
of TK activityas potential agents for the treatment of
cancer and other conditions.1 Several years ago, a group
at Pfizer reported that while the pyrimido[4,5-b]quino-
lin-4(1H)-one, 1, was a moderate Src kinase inhibitor,
the corresponding 5,10-dihydro analogue, 2, was 60-fold
more potent (IC50=32 and 0.5 mM respectively).2 Inter-
estinglywhen studiyng a series of EGFr inhibitors,
Parke-Davis (now Pfizer) found that the tricyclic analo-
gue 3 was more potent than the bicyclic analogue 4.3 We
had earlier shown that changing the 3-bromo aniline
substituent of 4 to 2,4-dichloro-5-methoxychanged the
kinase selectivity, with 5 now being a Src inhibitor.4 We
therefore envisioned that addition of the 2,4-dichloro-5-
methoxyaniline group present in 5 to the C-4 position of
1 and 2 might increase the Src inhibitoryactivityof these
compounds. Based on what was seen with 3 and 4, these
tricyclic analogues of 5 mayalso be more potent than the
parent bicyclic derivative.
vided the desired tricyclic derivative 6.6 Treatment of 2
with phosphorous oxychloride provided the intermediate
4-chloro derivative which was reacted with 2,4-dichloro-5-
methoxyaniline in 2-ethoxyethanol in the presence of
pyridine-HCl to provide the reduced analogue 7.
As shown in Table 1, 7 is more than 3000-fold more
potent in inhibiting Src kinase activitythan 6, with 7
having an IC50 of 3.3 nM in this assay.7 This dramatic
difference was also observed in cells, with 7 strongly
inhibiting the proliferation of Src transformed fibro-
blasts (IC50=190 nM). The disparities in the activities
seen with 6 and 7 are much more pronounced than those
observed byPfizer with the 4-one derivatives 1 and 2.
The preparation of the initial target compounds is shown
in Scheme 1. Treatment of 2-amino-6,7-dimethoxy-3-qui-
nolinecarbonitrile5 with dimethyl-formamide dimethyl
acetal gave the corresponding amidine. Subsequent addi-
tion of 2,4-dichloro-5-methoxyaniline in acetic acid pro-
Addition of a water solubilizing group to 7 would be
expected to improve the cell activityof this compound.
To this end, as shown in Scheme 2, a 2-methoxyethoxy
*Corresponding author. Tel.: +1-845-602-3567; fax: +1-845-602-
5561; e-mail: bosched@wyeth.com
0960-894X/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00628-0

2978
D. H. Boschelli et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2977–2980
Scheme 1. (a) (1) DMF-DMA; (2) 2,4-diCl-5-OMe aniline, AcOH;
(b) (1) POCl3; (2) 2,4-diCl-5-OMe aniline, pyridine-HCl, 2-ethoxy-
ethanol.
Table 1.
Scheme 2. (a) (1) 2-Chloroethyl methyl ether, K2CO3, DMF; (2) fum-
ing HNO3, dichloroethane; (b) (1) methyl cyanoacetate, piperidine,
MeOH; (2) Fe, AcOH; (c) HCONH2; (d) POCl3; (e) 2,4-diCl-5-OMe
aniline, tris(dibenzylidene)acetate, 2-dicyclohexylphosphino-20-(N,N-
dimethylamino)biphenyl, K3PO4, DME; (f) 2,4-diCl aniline, tris(di-
benzylidene)acetate, 2-dicyclohexylphosphino-20-(N,N-dimethylamino)-
biphenyl, K3PO4, DME.
Compd Ring
R1
R2
RAr Src enzyme7 Src cells7
(IC50 nM) (IC50 nM)
6
A
B
B
Me
Me
Me
Me
Me
OMe
OMe
>10,000
3.3
2.3
>10,000
190
28
7
12
CH2CH2– OMe
OCH3
13
15
16
18
B
A
B
CH2CH2–
OCH3
Me
Me
OMe
9.8
140
>10,000
120
CH2CH2– OMe
OCH3
CH2CH2–
OCH3
35% at
5 mM
16
The isomer of 12, namely 13, where the solubilizing
group is at C-7 was prepared in an analogous fashion
from 3-hydroxy-4-methoxybenzaldehyde. As for the
preparation of 12, the final product, 13, was obtained
via a palladium catalyzed coupling reaction.
Me
H
1.2
100
As shown in Table 1, 12, the analogue with the solubi-
lizing group at C-8, had about a 4-fold increase in both
Src enzymatic and cell activity compared to 13. For
comparision purposes, the analogue of 6 with a 2-
methoxyethoxy group at C-8 was prepared as shown in
Scheme 3. Intermediate 8 was reacted with mal-
ononitrile in the presence of b-alanine. Reduction with
iron in acetic acid provided 14. Formation of the ami-
dine derivative with dimethylformamide dimethyl acetal
followed byaddition of the aniline gave 15, the aro-
matic analogue of 12. As shown in Table 1, 15 was of
greatlyreduced activitycompared to 12.
group was added to C-7 and C-8 of 7 to provide 12 and
13, respectively. The 2-methoxyethoxy group is the
solubilizing group present on Tarceva, a 4-anilinoqui-
nazoline EGFr inhibitor currentlyundergoing clinical
trials.8
Alkylation of vanillin with 2-chloroethyl methyl ether
followed bynitration provided 8. Condensation of 8
with methyl cyanoacetate and subsequent reduction
with iron provided the bicyclic derivative 9. Cyclization
of 9 with formamide at high temperatures gave the
desired tricyclic core 10. Chlorination of 10 with phos-
phorous oxychloride provided 11. Attempts to add
2,4-dichloro-5-methoxyaniline under the standard con-
ditions of pyridine hydrochloride failed to provide the
desired product. It should be noted that although these
were the conditions used to prepare 7, the yield in this
reaction was only22%. Fortunately, implementation of
the palladium catalyzed coupling conditions reported by
Buchwald gave 12 in 64% yield.9
We earlier reported that while the optimal substituent
on the C-4 aniline was 2,4-dichloro-5-methoxy, the
2,4-dichloroaniline derivative also provided Src inhibi-
tion.4 Analogue 16 was prepared byaddition of
2,4-dichloroaniline to 11 under the palladium catalyzed
conditions used to prepare 12. When tested in the Src
enzyme and cell assays 16 was about 4-fold less potent
than 12. This corresponds to what was observed in a

D. H. Boschelli et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2977–2980
2979
Src cells, this compound had no effect on cortactin
phosphoryation in the Fyn cells.13
While the Src enzymatic activity of 12 is comparable to
that of 18, 12 is at least 3 times more potent in the Src
cell assay. However, in spite of the cellular potency and
Src selectivityof 12, this compound was not active when
tested in an in vivo nude mouse xenograft model
employing the Src-transformed fibroblasts.4 In attempts
to measure plasma levels in nude mice after administra-
tion of 12, no parent compound could be detected. It
was observed that while 12 was stable in the solid state,
solutions of 12 rather quicklyoxidized to give 15. The
investigation of more stable analogues of 12 is under-
way. We also hope to determine the reason for the
increased Src cell activityobserved with these analogues
and for their inactivityagainst Fyn.
Scheme 3. (a) (1) Malononitrile, b-alanine, MeOH; (2) Fe, AcOH;
(b) (1) DMF–DMA; (2) 2,4-diCl-5-OMe aniline, AcOH.
series of 3-quinolinecarbonitrile Src inhibitors where
17b was about 7-fold more potent than 17a when tes-
ted in the enzymatic assay (IC50s of 4.3 and 30 nM,
respectively).4
Acknowledgements
Optimization of 17b for cell activityand plasma levels
after oral dosing led to 18, which had an IC50 in the Src
cell assayof 100 nM (see Table 1).4 This compound was
4-fold less potent in inhibiting the proliferation of Fyn-
dependent cells (IC50 of 410 nM). Fyn is a Src family
kinase (SFK) and shares a high degree of structural
homologywith Src. Of all the additional 3-quinoline-
carbonitrile Src inhibitors we reported previously, we
did not observe greater than 10-fold selectivityfor Src
over Fyn.10ꢀ12 It was therefore verysurprising to find
that 12 did not inhibit the proliferation of Fyn-depen-
dent cells (IC50 >10 mM). Furthermore, no inhibition of
Fyn-dependent cell growth was observed with 7, 13, 15
or 16.
We thank Deanna Yackzo and JudyLucas for the
plasma level determination and in vivo study. We also
thank members of the Wyeth Discovery Analytical
Chemistrydepartment for the spectral data and ele-
mental analyses.
References and Notes
1. (a) Cohen, P. Nat. Rev. Drug Discov. 2002, 1, 309.
(b) Blume-Jensen, P.; Hunter, T. Nature 2001, 411, 355.
2. Dow, R. L.; Bechle, B. M.; Chou, T. T.; Goddard, C.;
Larson, E. Bioorg. Med. Chem. Lett. 1995, 5, 1007.
3. Rewcastle, G. W.; Palmer, B. D.; Bridges, A. J.; Showalter,
H. D. H.; Sun, L.; Nelson, J.; McMichael, A.; Kraker, A. J.;
Fry, D. W.; Denny, W. A. J. Med. Chem. 1996, 39, 918.
4. Boschelli, D. H.; Ye, F.; Wang, D. Y.; Dutia, M.; Johnson,
S.; Wu, B.; Miller, K.; Powell, D. W.; Arndt, K.; Discafani,
C.; Etienne, C.; Gibbons, J.; Grod, J.; Lucas, J.; Weber, J. M.;
Boschelli, F. J. Med. Chem. 2001, 44, 3965.
5. Althuis, T. H.; Kadin, S. B.; Czuba, L. J.; Moore, P. F.;
Hess, H. J. J. Med. Chem. 1980, 23, 262.
6. All compounds were characterized byMS, NMR and CHN
combustion analysis.
To further investigate this unpredicted finding, 12 was
tested for its abilityto inhibit the phosphorylation of
cortactin, a natural substrate for SFKs. As shown in
Figure 1, while 12 blocked the phosphoryation of tyr-
osine 421 of cortactin in a dose-dependent fashion in the
7. Compounds were tested according to the Src enzymatic,
cellular and Fyn cellular assays reported previously, see ref 4.
The IC50 values reported represent the means of at least 2
determinations with a typical variation of less than 40%
between replicate values.
8. Hidalgo, M.; Siu, L. L.; Nemunaitis, J.; Rizzo, J.; Ham-
mond, L. A.; Takimoto, C.; Eckhardt, S. G.; Tolcher, A.;
Britten, C. D.; Denis, L.; Ferrante, K.; Von Hoff, D. D.; Sil-
berman, S.; Rowinsky, E. K. J. Clin. Oncol. 2001, 19, 3267.
9. Old, D. W.; Wolfe, J. P.; Buchwald, S. L. A. J. Am. Chem.
Soc. 1998, 120, 9722.
10. Boschelli, D. H.; Wang, D. Y.; Ye, F.; Wu, B.; Zhang, N.;
Dutia, M.; Powell, D. W.; Wissner, A.; Arndt, K.; Weber,
J. M.; Boschelli, F. J. Med. Chem. 2001, 44, 822.
11. Wang, Y. D.; Miller, K.; Boschelli, D. H.; Ye, F.; Wu, B.;
Floyd, M. B.; Powell, D. W.; Wissner, A.; Weber, J. M.;
Boschelli, F. Bioorg. Med. Chem. Lett. 2000, 10, 2477.
12. Boschelli, D. H.; Wang, Y. D.; Ye, F.; Yamashita, A.;
Zhang, N; Powell, D.; Weber, J. M.; Boschelli, F. Bioorg.
Med. Chem. Lett. 2002, 12, 2011.
Figure 1. Selective inhibition of cortactin phosphorylation on tyrosine
421 by 12. Immunoblot of extracts from Src and Fyn transformed
Rat2 fibroblasts exposed to 12 for 5 h. Lysates from Src-transformed
(lanes 1–4), and Fyn-transformed (lanes 5–8) Rat2 fibroblast cells.
Lanes 1 and 5, no treatment; lanes 2 and 6, 0.1 mM 12; lanes 3 and 7,
0.5 mM 12; lanes 4 and 8, 1.0 mM 12.
13. Cell culture and immunoblots. Extracts from Src and Fyn

2980
D. H. Boschelli et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2977–2980
transformed Rat2 fibroblasts were prepared as previously
described.4 Compound 12 was freshlydissolved in DMSO
immediatelybefore addition to medium. Cells were exposed to
12 for 5 h. Lysates were analyzed on Novex 4–12% gradient
gels and transferred to PVDF membrane. Antibodies were
obtained from Upstate Biotechnology(anti-Cortactin), Bio-
Source (anti-pY421 Cortactin), and Chemicon International
(anti-Actin).