Design, synthesis, and evaluation of 2-anilino-4-(3,5-dicarboxamidespiperidine)-pyrimidines as anaplastic lymphoma kinase inhibitors

Full text of DOI:10.1002/bkcs.10454

Note
DOI: 10.1002/bkcs.10454
BULLETIN OF THE
M. Latif et al.
KOREAN CHEMICAL SOCIETY
Design, Synthesis, and Evaluation of 2-Anilino-4-
(3,5-dicarboxamidespiperidine)-pyrimidines as Anaplastic
Lymphoma Kinase Inhibitors
Muhammad Latif,†,‡ Sangjun Park,†,‡ Imran Ali,†,‡,§ Hyeonjeong Choe,†,‡ Chong Hak Chae,†,§
†,‡,
Chi Hoon Park,†,‡ Hyoung Rae Kim,† Chang-Soo Yun,†,‡, and Kwangho Lee
*
*
†Bio & Drug Discovery Division, Korea Research Institute of Chemical Technology, Daejeon 34114,
Republic of Korea. *E-mail: csyun@krict.re.kr; kwangho@krict.re.kr
‡Medicinal Chemistry & Pharmacology, Korea University of Science & Technology, Daejeon 34113,
Republic of Korea
§Korea Chemical Bank, Daejeon 34114, Republic of Korea
Received May 4, 2015, Accepted May 28, 2015, Published online August 27, 2015
Keywords: Anaplastic lymphoma kinase, ALK inhibitor, Pyrimidine derivatives, Kinase inhibitor,
Anticancer agent
Anaplastic lymphoma kinase (ALK), also known as CD246
(cluster of differentiation 246), is a receptor tyrosine kinase
(RTK) belonging to the insulin receptor superfamily.1 It
was first identified as a part of nucleophosmin-anaplastic lym-
phoma kinase (NPM-ALK) fusion protein derived from a
chromosomal translocation detected in patients with anaplas-
tic large-cell lymphoma (ALCL). NPM-ALK is a putative
oncogenic fusion protein of nucleophosmin (NPM) with the
intracellular domain of ALK and possesses a constitutively
active tyrosine kinase activity responsible for its oncogenic
effect through its aberrant activation of downstream signaling
including Akt, STAT3, and Erk1/2.2 Echinoderm microtu-
bule-associated protein-like 4-anaplastic lymphoma kinase
(EML4-ALK), a fusion of the ALK gene with echinoderm
microtubule-associated protein-like 4 (EML4), was identified
in Japanese patients with non-small-cell lung cancer (NSCLC)
during a crizotinib (Xalkori®, Pfizer) clinical study.3 Addi-
tional ALK-fusion proteins have recently been identified:
FN1-ALK is responsible for ovarian cancer, KIF5B-ALK
for NSCLC, and TPM3-ALK for inflammatory myofibroblas-
tic tumors (IMTs), diffuse large B-cell lymphoma (DLBCL),
and a variety of solid tumor types.4 In addition to ALK-fusion
genes, mutation-independent ALK overexpression has also
been associated with a poor prognosis in patients with
neuroblastoma.5
Among the various ALK inhibitors, the pyrimidine ALK
inhibitors NVP-TAE684 and LDK378 (ceritinib) retain
potency to many of crizotinib-resistant ALK mutants
(Figure 1).6 Of the many pyrimidine analogs, a new class of
pyrimidines, compound 1, which features C4-piperidinea-
mide substituents, has recently been reported by Lewis and
coworkers from Amgen.7 Compound 1 shows moderately
potent activity against ALK weight (IC50 = 0.17 μM) and
good selectivity over insulin-like growth factor-1 receptor
(IGF1R IC50 = 4.6 μM).7 Compound 1’s co-crystal structure
shows unusual hydrophobic interaction with the N-p-
methylbenzyl carboxamide group (PDB code 4DCE).7 The
carbonyl oxygen of the piperidine C3 amide has a hydro-
gen-bonding interaction with Lys1150, and the amide NH
interacts with Gly1269. The hydrophobic p-Me-benzyl has
a hydrophobic interaction with Phe1174, Ile1179, Phe1271,
Ile1171, and Ile1268 to compose its linear conformation, as
shown in Figure 1. This is quite different from the NVP-
TAE684 co-crystal structure, which possesses a U-shaped
binding conformation (PDB code 2XB7).8 We envisioned that
the piperidine-cis-3,5-dicarboxamides at the C4 pyrimidine
sitewouldretainthecompound 1-likehydrophobicinteraction
and the additional piperidine 5-amide can pick up an addi-
tional hydrogen-bonding interaction with Asp1270 (vide
infra). This additional hydrogen bonding to Asp1270 may
relieve the dependence on the hydrophobic interaction of
the compound 1 and improve the Karpas-299 cellular potency
without a large shift in cellular activity. To prove this hypoth-
esis, we decided to investigate the C4-(piperidine-cis-3,5-
dicarboxamides)-pyrimidines as inhibitors of ALK. To avoid
stereochemistry issues, symmetric bisamide analogs were pre-
pared, as depicted in Figure 2.
A general synthetic chemistry toward 2-anilino-4-(3,5-
dicarboxamidespiperidine)-pyrimidines is described in
Scheme 1. 3,5-Pyridinedicarboxylic acid 2 was esterified
using thionyl chloride in methanol to give 3. Hydrogenation
(PtO2/AcOH) gave a cis/trans mixture of 4, which was then
protected with di-tert-butyl dicarbonate to give 5. The N-
Boc-protected 5 cis isomer was selectively precipitated from
the N-Boc-protected cis/trans mixture in hexane after room
temperature, giving a white solid.9 cis-5 was treated with
hydrogen chloride in dioxane to provide a hydrochloride salt
of cis-4, whichwas thencoupledto2,4,5-trichloropyrimidines
in the presence of triethylamine in isopropanol to afford 6 in a
regioselective manner for Scaffold 1 series (vide infra). Cou-
pling of 6 with 2,3,4-trimethoxyaniline in dimethyl sulfoxide
smoothly provided 7. Hydrolysis using lithium hydroxide,
Correction added on 01 October 2015, after first online publication: ISSN (Print) has been corrected.
Bull. Korean Chem. Soc. 2015, Vol. 36, 2397–2400
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Wiley Online Library
2397

Note
BULLETIN OF THE
ISSN (Print) 0253-2964 | (Online) 1229-5949
KOREAN CHEMICAL SOCIETY
Figure 1. ALK inhibitors.
Scheme 1. 2-Anilino-4-(3,5-dicarboxamidespiperidine)-pyrimi-
dines preparation.
ALK at the biochemical and cellular level. However, the
methyl, ethyl, i-propyl, benzyl analogs 10, 12, 14, 15, 17,
19, 21, 22 were inactive in our assay (both >1 μM in enzyme
assay and >10 μM in cell-based assay). The n-butyl analog is
not as potent as n-propyl analogs in a relevant study (data not
shown). Therefore, their ALK biochemical and cellular activ-
ities depend on the amide tail length, and the n-propyl tail deri-
vatives are most potent in this study. Since the n-propyl tail
C4-(piperidine-cis-3,5-dicarboxamides)-pyrimidine 13 is
identified as potent to ALK, we compared the dicarboxamide
13 to compound 1-like mono-n-propyl tail carboxamide 23.13
Compound 23 was far less active, showing little ALK weight
inhibition at 1 μM and >10 μM IC50 in cell-based assay. With
this direct comparison, the C4-(piperidine-cis-3,5-dicarboxa-
mides)-pyrimidines proved to be superior to its mono-carbox-
amides. The bis-N,N-dimethyl carboxamide derivatives 11
and 18 were both inactive at 10 μM in the biochemical assay.
We tried to explain this n-propyl tail effect to ALK inhibi-
tion with a modeling study.14 Analysis of ALK residues inter-
acting with the compounds reveals information valuable for
the evaluation of binding affinity. The strong interaction of
aminopyridine with ALK is attributed to the two hydrogen
bonds between the 2-aminopyridine moiety and the backbone
NH and carbonyl oxygen of the Met1199 hinge residue. The
trimethoxyphenyl at the 2-amino position is sandwiched
between the narrow groove of Leu1122 and Gly1202 of the
hinge region. The 5-chlorine of the pyrimidine makes hydro-
phobic (halogen) interaction with the Leu1196 gate keeper
residue (Figure 3). Additional stabilization of compound 13
to ALK comes from the interaction of the symmetric side
Figure 2. Postulated hydrogen bonding and hydrophobic interaction
on the tails.
followed by final 1-ethyl-3-(3-dimethylaminopropyl)carbo-
diimide (EDCI)-mediated amide formation with various
amines gave the desired analogs of 8. The final products were
purified using silica gel column chromatography and reverse-
phase HPLC if necessary. The C5-H-pyrimidine analogs
(Scaffold 2) were prepared in similar fashion.
Since compound 1 displays a unique structure–activity rela-
tionship (SAR) depending on the hydrophobic interaction of
the benzylamide substitution, we surveyed the bis-amide sub-
stitution effect including NH2, mono-substituted amides
(methyl, ethyl, n-propyl, i-propyl, benzyl), and a di-substi-
tuted amide (dimethyl), as summarized in Table 1. The bis-
amide substitution SAR study was conducted on the C5-chlor-
opyrimidine ring with 2,3,4-trimethoxyaniline attached on the
C2 position (Scaffold 1) and the C2-N-(3,4,5-trimethoxyani-
lino)pyrimidine (Scaffold 2), which was used in the study of
compound 1 analogs by the Amgen group.7
Compounds 9–22 were evaluated in an ALK % inhibition
assay at 10 μM and 1 μM. Compounds that showed above
50% ALK inhibition at1 μM were analyzed for ALK IC50
values.10 Compounds 9, 13, and 20 possess moderate cellular
H3122 potencies (IC50s = 3.7–7.2 μM) (H3122: an EML4-
ALK addicted NSCLC cell).11 Compound 9 has an unsubsti-
tuted bis-amide tail (NH2) and possesses a lower cLogP (1.40)
than compound 1 (4.09).12 It is interesting to note that com-
pounds 13 and 20, which have n-propyl tails, are active against
Bull. Korean Chem. Soc. 2015, Vol. 36, 2397–2400
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de 2398

BULLETIN OF THE
Note
KOREAN CHEMICAL SOCIETY
Table 1. SAR table of C4-(piperidine-cis-3,5-dicarboxamides)-pyrimidines
ALK wt % inhibition
ALK weight % inhibition
ALK weight
CP H3122
Scaffold
R1
R2
@ 10 μM
@ 1 μM
IC50 (μM)
IC50 (μM)
Crizotinib
Compound 1
9
—
—
99
55
0
28
100
0
1.4
66
0
0
4
—
—
58
0
0
0
66
0
0
3.5
8
0
0.036
0.174a
0.64
—
0.28
—
1
1
1
1
1
1
1
2
2
2
2
2
2
2
H
Me
Me
Et
n-Pr
i-Pr
Bn
H
H
Me
H
H
H
H
H
H
Me
H
H
H
4.7
>10
>10
>10
3.7
>10
>10
—
10
11
12
13
14
15
16
17
18
19
20
21
22
23
—
—
0.58
—
—
H
—
Me
Me
Et
n-Pr
i-Pr
Bn
—
—
—
—
6
—
—
99
6
56
65
83
10
0
0.11
—
7.2
—
H
—
—
1
—
>10
— Not available.
a Data from Ref. 7.
additional hydrophobic interactions with the extended hydro-
phobic pockets formed by the shift-out movement of Phe1271
from the hydrophobic pocket, as shown in the crystal structure
4DCE (DFG shifted conformation).7 Therefore, one strategy
to improve the binding affinity would be to add a long hydro-
phobic substituent (substituted phenyl, substituted cyclo-
hexyl, etc.) at the C3 site, which would make favourable the
hydrophobic interactions. The NH of the amide bond at the
piperidine C5 position forms hydrogen bond with the carboxyl
of the Asp1270 residue. The n-propyl moiety at the piperidine
C5 amide shows no specific interactions. Overall, the piperi-
dine 3-substituents of compound 13 are located in the hydro-
phobic region and the 5-substituents are exposed to the solvent
region. Therefore the hydrophobicity of the 3-substituents of
the piperidine likely increases ALK potency, whereas the 5-
position may prefer hydrophilic substituents. In this regard,
the asymmetric bis-amidopiperidines are of interest for
ALK inhibition. In compounds 11 and 18, the C3 and C5
piperidine-amide (NH) hydrogen-bonding interactions to
Gly1269 and Asp1270 carbonyls are blocked by methyl sub-
stituents, resulting in loss of activity.
Figure 3. Modeling of compound 13 (green) to ALK wt and compar-
ison to compound 1 (thin orange lines).
chainsattheC3andC5positionsofthepiperidineandresidues
near the aspartate-phenylalanine-glycine (DFG) region of
ALK.15 The amide bond at piperidine C3 position of com-
pound 13 forms two hydrogen bonds: the carbonyl oxygen
to the NH2 of Lys1150 side chain, and the amide NH to the
main chain carbonyl oxygen of Gly1269, as depicted in
Figure 3. The n-propyl of the piperidine C3 amide makes
In conclusion, inspired by a previous report by Lewis and
coworkers, we hypothesized an additional binding interaction
with an additional amide to Asp1270 as beneficial for ALK
Bull. Korean Chem. Soc. 2015, Vol. 36, 2397–2400
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de 2399

BULLETIN OF THE
Note
KOREAN CHEMICAL SOCIETY
T. Mitsudomi, Y. Tanio, H. Mano, N. Engl. J. Med. 2010,
363, 1734.
7. M. C. Bryan, D. A. Whittington, E. M. Doherty, J. R. Falsey,
A. C. Cheng, R. Emkey, R. L. Brake, R. T. Lewis, J. Med. Chem.
2012, 55, 1698.
8. R. T. Bossi, M. B. Saccardo, E. Ardini, M. Menichincheri,
L. Rusconi, P. Magnaghi, P. Orsini, N. Avanzi, A. L. Borgia,
M. Nesi, T. Bandiera, G. Fogliatto, J. A. Bertrand, Biochemisty
2010, 49, 6813.
9. The proton NMRs of the 5 cis/trans isomers are attached to Sup-
plement data and consistent to previous report:I. Collins,
W. B. Davey, M. Rowley, K. Quirk, F. A. Bromidge,
R. M. McKernan, S.-A. Thompson, K. A. Wafford, Bioorg.
Med. Chem. Lett. 2000, 10, 1381.
10. The kinase inhibition assay for ALK weight was conducted with
homogeneous time-resolved fluorescence (HTRF) KinEASE-
TK assay from Cisbio according to the manufacturer’s instruc-
tions. PurifiedALKandinsulinreceptorproteinswerepurchased
fromCarna Biosciences(Kobe, Japan). Kinase, biotinylatedsub-
strate peptide, and inhibitors were added for pre-incubation in
384-well plates, and then the reaction was started by the addition
of ATP. After 30 min, an antiphosphotyrosine antibody labeled
with europium cryptate and streptavidin labeled with the fluoro-
phore XL665 were added to the reaction mixtures. The FRET
between europium cryptate and XL665 was measured to quan-
tify the phosphorylation of the substrate peptide. An Envision
multilabel reader (Perkin Elmer, Waltham, MA, USA) was used
tomeasurethefluorescenceofthesamplesat620 nm(Eu-labeled
antibody) and 665 nm (XL665-labeled streptavidin) after excita-
tion at 320 nm. The ratio of both intensities was calculated with
12 different inhibitor concentrations (0.03–10 000 nM), which
was plotted against inhibitor concentrations to determine IC50
values. The Prizm program (La Jolla, CA, USA) was used for
IC50 determination.
11. Cells were seeded in 96-well plates in a medium with 10% FBS.
The next day, cells were treated with different concentrations
(0.64–10 000 nM) of compounds for 72 h at 37 ꢀC. Then, cells
were fixed for SRB (sulfo-rhodamine B) assay. The results were
quantified by spectrophotometry at 520 nm. The Prizm program
was used for IC50 determination.
12. Compound 9 was further evaluated for kinase % activity over
100 other kinases at 1 μM concentration (Ambit assay). Only
three(ACK1, ALK,FMS)kinaseactivitiesgotmoderatelyinhib-
ited by compound 9. Ambit assay results of compounds 9 are
given as Supporting Information.
binding and could relieve the cellular activity shift presumably
due to the hydrophobic tail.7 The additional amide interaction
contributes marginally to its ALK binding, and many of the
C4-(piperidine-cis-3,5-dicarboxamides)-pyrimidines possess
little to modest potency. Interestingly, the n-propyl tail con-
taining analogs 13 and 20 show a level of ALK activity, and
we tried to explain this observation with a modeling study.
Although the symmetrical bisamides 9, 13, and 20 described
in this communication showed moderate biochemical and cel-
lular activities, they are much less potent compared to the ref-
erence (crizotinib).
Acknowledgments. This work was supported by the Korea
Research Institute of Chemical Technology (SI-1506), the
Korea Chemical Bank (SI-1512-01), and Ministry of Science,
ICT
and
Future
Planning
of
Korea
(NRF-
2012M3A9A9054902).
Supporting Information. Analytics of compounds 9, 13, 20
and Ambitassay result ofcompound9are attached asSupport-
ing Information.
References
1. R. H. Palmer, E. Vernersson, C. Grabbe, B. Hallberg,
Biochem. J. 2009, 420, 345.
2. A. V. Galkin, J. S. Melnick, S. Kim, T. L. Hood, N. Li, L. Li,
G. Xia, R. Steensma, G. Chopiuk, J. Jiang, Y. Wan, P. Ding,
Y. Liu, F. Sun, P. G. Schultz, N. S. Gray, M. Warmuth, Proc.
Natl. Acad. Sci. U. S. A. 2007, 104, 270.
3. M. Soda, Y. L. Choi, M. Enomoto, S. Takada, Y. Yamashita,
S. Ishikawa, S.-i. Fujiwara, H. Watanabe, K. Kurashina,
H. Hatanaka, M. Bando, S. Ohno, Y. Ishikawa, H. Aburatani,
T.Niki,Y.Sohara,Y.Sugiyama,H.Mano,Nature2007,448,561.
4. Fibronectin1-ALK (FN1-ALK); kinasin family member 5B-
ALK (KIF5B-ALK); tropomyosin3-ALK (TPM3-ALK). For
detail,(a) J. P. Koivunen, C. Mermel, K. Zejnullahu,
C. Murphy, E. Lifshits, A. J. Holmes, H. G. Choi, J. Kim,
D. Chiang, R. Thomas, J. Lee, W. G. Richards,
D. J. Sugarbaker, C. Ducko, N. Lindeman, J. P. Marcoux,
J. A. Engelman, N. S. Gray, C. Lee, M. Meyerson,
P. A. Jänne, Clin. Cancer Res. 2008, 14(13), 4275;
(b) T. Sasaki, S. J. Rodig, L. R. Chirieac, P. A. Jänne, Eur.
J. Cancer 2010, 46, 1773; (c) P. Minoo, H.-Y. Wang, Int.
J. Clin. Exp. Pathol. 2012, 5(5), 397; (d) H. Ren, Z.-P. Tan,
X. Zhu, K. Crosby, H. Haack, J.-M. Ren, S. Beausoleil,
A. Moritz, G. Innocenti, J. Ruch, Y. Zhang, X.-M. Zhou, T.-
L. Gu, Y.-F. Yang, M. J. Comb, Cancer Res. 2012, 72, 3312.
5. B. Hallberg, R. H. Palmer, Nat. Rev. Cancer 2013, 13, 685.
6. (a) A. T. Shaw, B. Solomon, Clin. Cancer Res. 2011, 17, 2081;
(b) J. J. Cui, M. Tran-Dubé, H. Shen, M. Nambu, P. Kung,
M. Pairish, L. Jia, J. Meng, L. Funk, I. Botrous, M. McTigue,
N. Grodsky, K. Ryan, E. Padrique, G. Alton, S. Timofeevski,
S. Yamazaki, Q. Li, H. Zou, J. Christensen, B. Mroczkowski,
S. Bender, R. S. Kania, M. P. Edwards, J. Med. Chem. 2011,
54, 6342; (c) Y. L. Choi, M. Soda, Y. Yamashita, T. Ueno,
J. Takashima, T. Nakajima, Y. Yatabe, K. Takeuchi,
T. Hamada, H. Haruta, Y. Ishikawa, H. Kimura,
13. Compound23waspreparedusingasimilarchemistryasreported
in Ref. 7.
14. Modeling studies were performed using the Glide software con-
tained in the Maestro 9.1 software (Portland, OR, USA) with
standard-precision (SP) options. [R. A. Friesner, J. L. Banks,
R. B. Murphy, T. A. Halfren, J. J. Klicic, D. T. Mainz,
M. P. Repasky, E. H. Knoll, M. Shelley, J. K. Perry,
D. E. Shaw, P. Francis, P. S. Shenkin, J. Med. Chem. 2004,
47, 1739]. The crystal structure of ALK (PDB code: 4DCE from
Ref. 6 was used with structural defects fixed by the Protein Prep-
aration Wizard module. The bound ligand conformation was
generated with a good root mean square displacement
(RMSD) of 0.3 A.
15. R. Roskoski Jr., Pharmacol. Res. 2013, 68, 68.
Bull. Korean Chem. Soc. 2015, Vol. 36, 2397–2400
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de 2400