Pyrimido[4,5-d]pyrimidin-4(1H)-one derivatives as selective inhibitors of EGFR threonine790 to methionine790 (T790M) mutants

Article abstract of DOI:10.1002/anie.201302313

Catching the mutants: Pyrimido[4,5-d]pyrimidin-4(1H)-one derivatives (see example) were identified as specific inhibitors of EGFR^T790M mutants. The compounds bound with T790M or L858R/T790M mutants with significantly lower K_d values

Full text of DOI:10.1002/anie.201302313

Angewandte
Chemie
DOI: 10.1002/anie.201302313
Drug Design
Pyrimido[4,5-d]pyrimidin-4(1H)-one Derivatives as Selective
Inhibitors of EGFR Threonine790 to Methionine790 (T790M) Mutants**
Tianfeng Xu, Lianwen Zhang, Shilin Xu, Chao-Yie Yang, Jinfeng Luo, Fang Ding, Xiaoyun Lu,*
Yingxue Liu, Zhengchao Tu, Shiliang Li, Duanqing Pei, Qian Cai, Honglin Li, Xiaomei Ren,
Shaomeng Wang, and Ke Ding*
The epidermal growth factor receptor (EGFR, erbB1, HER1)
has been well-validated as a molecular target in anticancer
drug discovery. In non-small-cell lung cancer patients
(NSCLCs) harboring active mutations in the EGFR tyrosine
kinase domain (L858R and del E746-A750),[1–4] the first
generation inhibitors, gefinitib and erlotinib, have achieved
significant clinical benefits but emerging acquired resistance
to them has become a major clinical challenge. The “gate-
keeper” T790M mutation (threonine790!methionine790) of
EGFR, by which the binding of ATP with the kinase is
favored,[5] is one of the primary mechanisms for resistance
and plays a role in the circa 50% of NSCLC patients who
acquired clinical resistance.[6,7] Although the Cys797-chelat-
ing irreversible EGFR inhibitors displayed promising poten-
tial to overcome EGFRT790M related resistance in animal
models, their non-selective inhibition against wild-type
EGFR (EGFRWT) and/or other kinases results in a relatively
low maximal-tolerated-dose (MTD) and poor clinical out-
comes in human patients.[8–11] Inhibitors selectively targeting
EGFRT790M mutants are an attractive strategy for the clinical
management of NSCLC patients with acquired resistance.
However, because EGFRWT and the EGFRT790M mutants
share highly similar three-dimensional structures and have
almost identical binding affinities with ATP,[5] nearly all of the
reported irreversible EGFR inhibitors displayed equal poten-
cies against the T790M mutants and the wild-type enzyme,
highlighting the challenge in the search for EGFRT790M
mutant-selective inhibitors. Only recently, WZ4002 was
reported as a new irreversible EGFR inhibitor displaying
moderate selectivity on EGFRT790M mutants over the
wild-type kinase.[12] A phase I clinical trial was recently
initiated on another moderately mutant-selective EGFR
inhibitor CO-1686 (Kd(EGFRWT)/Kd(EGFRL858R/T790M) = 25,
NCT01526928) whose chemical structure was not disclosed,[13]
and PKC412 was reported as a novel reversible T790M
mutant-selective EGFR inhibitor with promising in
vivo efficacy.[14] Herein, we wish to report the successful
discovery of novel pyrimido[4,5-d]pyrimidin-4(1H)-one-
based EGFRT790M inhibitors with more than 100-fold selec-
tivity over the wild-type kinase.
We have successfully designed compounds 1 and 2
(Scheme 1) as novel EGFRT790M inhibitors with low nano-
molar IC50 and Kd values. However, these compounds only
displayed four-fold selectivity on EGFRT790M mutants over
EGFRWT [15a,b]
The use of conformational constraint is a gen-
.
eral strategy with which to improve ligand selectivity for
a molecular target, and accordingly a series of pyrimido[4,5-
d]pyrimidin-4(1H)-one derivatives 3a–3h with more rigid
conformations based on the structure-activity relationship
(SAR) studies of compounds 1 and 2 (Scheme 1)[15a,b] were
designed. The compounds were readily synthesized by using
the similar procedures to that of 3a (Scheme 2; Supporting
Information, Scheme S1). Briefly, a direct nucleophilic cou-
pling of commercially available ethyl 2,4-dichloropyrimidine-
5-carboxylate (4) with tert-butyl-3-aminophenylcarbamate (5)
produced ethyl 4-[{3-[(tert-butoxycarbonyl)amino]phenyl}a-
mino]-2-chloropyrimidine-5-carboxylate (6).[16] Hydrolysis of
compound 6 with 1m NaOH in a H2O/THF mixed solution
yielded the carboxylic acid (7). The condensation of 7 and 8a
in the presence of HATU and DIPEA in dry CH2Cl2 gave the
intermediate 9a. Compound 9a was coupled with aniline by
nucleophilic substitution followed by deprotection with 50%
trifluoroacetic acid in CH2Cl2 to yield the key precursor 10a.
The new inhibitor 3a was finally obtained by acryloylation of
10a with acryloyl chloride.
[*] T. Xu,[+] L. Zhang,[+] S. Xu, J. Luo, F. Ding, X. Lu, Y. Liu, Z. Tu,
Prof. D. Pei, Q. Cai, X. Ren, Prof. K. Ding
Institute of Chemical Biology, Guangzhou Institutes of Biomedicine
and Health, Chinese Academy of Sciences
190 Kaiyuan Avenue, Guangzhou 510530 (China)
E-mail: ding_ke@gibh.ac.cn
T. Xu,[+] S. Xu
University of Chinese Academy of Sciences
19 Yuquan Road, Beijing 100049 (China)
Dr. C.-Y. Yang, Prof. S. Wang
The University of Michigan Comprehensive Cancer Center
Ann Arbor, MI 48109 (USA)
S. Li, Prof. H. Li
East China University of Science and Technology
130 Meilong Road, Shanghai 200237 (China)
[+] These authors contributed equally to this work.
[**] We thank National Basic Research Program of China (no
2010CB529706, 2009CB940904), National Natural Science Foun-
dation (no 21072192, 21102146), Key Project on Innovative Drug of
Guangdong Province (no 2011A080501013), and Key Project on
Innovative Drug of Guangzhou City (no 2009Z1-E911, 2010J-E551)
for their financial support.
The binding affinities of the compounds with EGFRWT
and its T790M mutants were determined with an active-site-
dependent competition binding assay conducted by Ambit
Bioscience, San Diego, USA. The kinase inhibitory activities
of the compounds were also evaluated by the well-established
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201303213.
Angew. Chem. Int. Ed. 2013, 52, 8387 –8390
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8387

.
Angewandte
Communications
mutants, it dramatically impacted their effects on wild-type
EGFR. For instance, compounds 3a–3d bound to EGFR L858R/
T790M
with Kd values of 3.6, 1.1, 0.88, and 1.8 nm, respectively,
and they also inhibited the enzymatic function of EGFRL858R/
T790M
with similar IC50 values (7.03, 2.09, 1.69, and 1.52 nm,
respectively). However, their binding affinities with EGFRWT
were 210, 72, 17 and 140 nm, respectively; that is, 58.3, 65.5,
19.3, and 77.8-fold greater than the Kd values with EGFRL858R/
T790M. The most rigid compounds (3a and 3d) were indeed
significantly more selective than the other compounds. The
tetracyclic compound 3e displayed almost equal potencies
and selectivity to that of 3d on EGFRT790M mutants, but its
poor solubility impeded further investigation. Although
compound 3 f, with an N- morpholino moiety, was 2–3 times
more selective than 3d, its potency was 3–5-fold lower.
Compound 3h was more selective but slightly less potent than
compound 3d, while compound 3g with a 4-(dimethylami-
no)piperidin-1-yl substituent displayed improved selectivity
and equal potency to that of 3d (Table 1). Further kinase
profiling investigations against a panel of 456 kinases revealed
that 3g displayed excellent selectivity on EGFRT790M mutants
at 100 nm, about 77- and 32-times greater than its Kd values
with EGFRT790M and EGFRL858R/T790M, respectively. Only four
of the 395 non-mutate kinases, CAMK2A, DAPK2, DAPK3,
and HIPK2, manifested binding with compound 3g (inhib-
ition rate ꢀ 65%, or Ctrl% ꢁ 35%), and the S(35) selectivity
score of 3g was 0.01. To the best of our knowledge, compound
3g is one of the most selective EGFRT790M inhibitors reported
to date (Supporting Information, Figures S7, S8).
The strong EGFRT790M kinase inhibition of the new
inhibitors was further confirmed by examination of the effects
of the representative compounds 3d and 3g on the activation
of EGFR and the downstream signals in gefitinib-resistant
H1975 NSCLC cells harboring EGFRL858R/T790M. It was shown
that compounds 3d and 3g dose-dependently inhibited the
phosphorylation of EGFR and decreased the protein levels of
downstream p-Akt, p-Erk, while gefitinib showed no activity
(Figure 1). Further studies revealed that the compounds also
potently inhibited EGFR activation in NCI-H820 NSCLC
cells with EGFRdel E746-E749/T790M, but had no obvious effect in
a panel of NSCLC cells with wild-type EGFR (that is, NCI-
H1299, A549, 95D, NCI-H446, NCI-H358, NCI-H661, NCI-
Scheme 1. Knowledge-based design of pyrimido[4,5-d]pyrimidin-4(1H)-
one derivatives 3a–3h as new EGFRT790M inhibitors.
FRET-based Z’-Lyte assay (Table 1).[15a,b] All four of the rigid
compounds (3a–3d) bound effectively to EGFRT790M and
L858R/T790M
EGFR
with Kd values in the low-nanomolar range.
The compounds also potently inhibited the kinase functions
with IC50 values between 1.52 nm and 13.87 nm. Although the
conformational rigidity of the compounds apparently had no
effect on the binding affinities and kinase inhibitory activities
of the compounds on EGFRT790M and EGFRL858R/T790M
Scheme 2. Synthesis of compound 3a. DIPEA=diisopropylethylamine, HATU=O-(7-azabenzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium hexa-
fluorophosphate, TFA=trifluoroacetic acid.
8388
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 8387 –8390

Angewandte
Chemie
Table 1: Binding and kinase inhibitory activities of the new EGFRT790M inhibitors.
No.
EGFR Binding affinity (Kd, [nm])[a]
Kinase inhibition (IC50 [nm])[b]
WT
T790M
L858R/T790M
WT:T790M
WT:DM
T790M
L858R/T790M
3a
3b
3c
3d
3e
3 f
3g
3h
1
210
72
17
140
110
950
310
420
1.8
3.6
1.1
0.88
1.8
1.1
6.1
2.6
3.1
0.3
116.7
81.8
39.5
58.3
65.5
19.3
77.8
100
155.7
119.2
135.5
4.0
13.87ꢂ5.29
3.22ꢂ0.49
3.36ꢂ0.95
4.50ꢂ1.33
3.57ꢂ2.47
14.84ꢂ0.31
4.55ꢂ0.25
8.41ꢂ0.40
0.67
7.03ꢂ2.85
2.09ꢂ0.41
1.69ꢂ0.51
1.52ꢂ0.81
1.82ꢂ0.20
7.00ꢂ0.19
2.18ꢂ0.1
3.85ꢂ0.27
0.93
0.88
0.43
1.4
0.69
4
1.3
1.3
0.29
100
159.4
237.5
238.5
323.1
4.1
1.2
[a] Binding constant values (Kd) were determined from Ambit KINOMEscan. The data are means from two independent experiments. [b] EGFR kinase
assays were performed using the FRET-based Z’-Lyte assay according to the manufacturer’s instructions. The compounds were incubated with the
kinase reaction mixture for 1.5 h before measurement. The data are means from at least three independent experiments. WT=wild-type. DM=L858R/
T790M double mutations.
Table 2: The anti-proliferative activities of 3d and 3g against cells with
different EGFR.[a]
Cells
EGFR status
Anti-proliferation (IC50, [mm])
3d
3g
H1975
H322
L858R/T790M
0.143ꢂ0.026
>30
0.086ꢂ0.018
>30
WT
A549
WT
WT
WT
WT
8.85ꢂ0.76
>30
>30
>30
>30
H1299
H1703
H661
>30
20.15ꢂ9.92
20.83ꢂ2.95
>30
95D
WT
>30
H358
WT
del E746-A750
WT
6.25ꢂ1.15
0.039ꢂ0.013
9.96ꢂ5.08
3.96ꢂ0.11
>30
HCC827
HL-7702
HLF-1
0.049ꢂ0.027
>30
WT
10.47ꢂ1.46
[a] The anti-proliferative activities of the compounds were evaluated
using MTS assay. The data were means from at least four independent
experiments.
Figure 1. Inhibition by compounds 3d and 3g of the activation of
EGFR and downstream signaling in H1975 NSCLC cells harboring
EGFRL858R/T790M. In contrast, gefitinib (Gfb) has no effect (right).
a view to overcoming the clinically acquired resistance against
gefitinib.
H1703, NCI-H322 cancer cells; Supporting Information,
Figure S2).
In summary, we have described the discovery of a new
series of pyrimido[4,5-d]pyrimidin-4(1H)-one derivatives that
are highly selective and potent inhibitors targeting the
resistance of related EGFRT790M mutants. The extraordinary
selectivity of the compounds over wild-type EGFR and other
kinases makes them attractive lead compounds for future
drug development.
The cell-growth inhibitory effects of the compounds were
also investigated against a panel of NSCLC cells with
different EGFR status (Table 2; Supporting Information,
Table S1). Compounds 3d and 3g potently inhibited the
growth of H1975 NSCLC cells harboring the EGFR T790M
mutation with IC50 = 0.143 and 0.086 mm, respectively, but
were 60–400-fold less potent to NSCLC cells with wild-type
EGFR. These results correlated exactly with the selective
inhibition by 3d and 3g of EGFRT790M mutants. Compounds
3d and 3g were also significantly less cytotoxic to normal HL-
7702 (normal human liver cells) and HLF-1 (human lung
fibroblast cells) harboring EGFRWT. Further investigation
revealed that the compounds dose-dependently arrested the
H1975 cells in the G1 phase (Supporting Information, Fig-
ure S3), caused cell apoptosis (Supporting Information, Fig-
ure S4), and strongly inhibited tumorigenesis (Supporting
Information, Figure S5) and metastasis of H1975 cells in vitro
(Supporting Information, Figure S6). Thus they may be used
as promising lead compounds for further drug discovery with
Received: March 19, 2013
Published online: June 26, 2013
Keywords: drug design · inhibitors · selectivity · T790M mutants
.
[1] T. S. Mok, Y. L. Wu, S. Thongprasert, C. H. Yang, D. T. Chu, N.
Saijo, P. Sunpaweravong, B. H. Han, B. Margono, Y. Ichinose, Y.
Nishiwaki, Y. Ohe, J. J. Yang, B. Chewaskulyong, H. Jiang, E. L.
Duffield, C. L. Watkins, A. A. Armour, M. Fukuoka, N. Engl. J.
Med. 2009, 361, 947 – 957.
[2] R. Rosell, T. Moran, C. Queralt, R. Porta, F. Cardenal, C. Camps,
M. Majem, G. Lopez-Vivanco, D. Isla, M. Provencio, A. Insa, B.
Massuti, J. L. Gonzalez-Larriba, L. Paz-Ares, I. Bover, R.
Garcia-Campelo, M. A. Moreno, S. Catot, C. Rolfo, N. Reguart,
Angew. Chem. Int. Ed. 2013, 52, 8387 –8390
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8389

.
Angewandte
Communications
R. Palmero, J. M. Sanchez, R. Bastus, C. Mayo, J. Bertran-
Alamillo, M. A. Molina, J. J. Sanchez, M. Taron, S. L. C. Grp, N.
Engl. J. Med. 2009, 361, 958 – 967.
[11] A. Campbell, K. L. Reckamp, D. R. Camidge, G. Giaccone, S. M.
Gadgeel, F. R. Khuri, J. A. Engelman, L. J. Denis, J. P. OꢁCon-
nell, P. A. Janne, J. Clin. Oncol. 2010, 28, 7596 (suppl; abstr).
[12] W. J. Zhou, D. Ercan, L. Chen, C. H. Yun, D. N. Li, M. Capelletti,
A. B. Cortot, L. Chirieac, R. E. Iacob, R. Padera, J. R. Engen,
K. K. Wong, M. J. Eck, N. S. Gray, P. A. Janne, Nature 2009, 462,
1070 – 1074.
[13] A. O. Walter, R. Tjin, H. Haringsma, K. Lin, A. Dubrovskiy, K.
Lee, T. S. Martin, R. Karp, Z. Zhu, D. Nu, M. Nacht, K. Suda, T.
Mitsudomi, C. R. C. Russell Petter, W. F. Westlin, J. Singh, M.
Raponi, A. Allen, Cancer Res. 2012, 72 (8 Supplement), S1791.
[14] H.-J. Lee, G. Schaefer, T. P. Heffron, L. Shao, X. Ye, S. Sideris, S.
Malek, E. Chan, M. Merchant, H. La, S. Ubhayakar, R. L.
Yauch, V. Pirazzoli, K. Politi, J. Settleman, Cancer Discovery
2013, 3, 168 – 181.
[15] a) S. H. Chang, L. W. Zhang, S. L. Xu, J. F. Luo, X. Y. Lu, Z.
Zhang, T. F. Xu, Y. X. Liu, Z. C. Tu, Y. Xu, X. M. Ren, M. Y.
Geng, J. Ding, D. Q. Pei, K. Ding, J. Med. Chem. 2012, 55, 2711 –
2723; b) S. L. Xu, L. W. Zhang, S. H. Chang, J. F. Luo, X. Y. Lu,
Z. C. Tu, Y. X. Liu, Z. Zhang, Y. Xu, X. M. Ren, K. Ding,
MedChemComm 2012, 3, 1155 – 1159.
[16] H. Hisamichi, R. Naito, A. Toyoshima, N. Kawano, A. Ichikawa,
A. Orita, M. Orita, N. Hamada, M. Takeuchi, M. Ohta, S. I.
Tsukamoto, Bioorg. Med. Chem. 2005, 13, 4936 – 4951.
[3] C. H. Yun, T. J. Boggon, Y. Q. Li, M. S. Woo, H. Greulich, M.
Meyerson, M. J. Eck, Cancer Cell 2007, 11, 217 – 227.
[4] K. D. Carey, A. J. Garton, M. S. Romero, J. Kahler, S. Thomson,
S. Ross, F. Park, J. D. Haley, N. Gibson, M. X. Sliwkowski,
Cancer Res. 2006, 66, 8163 – 8171.
[5] C. H. Yun, K. E. Mengwasser, A. V. Toms, M. S. Woo, H.
Greulich, K. K. Wong, M. Meyerson, M. J. Eck, Proc. Natl.
Acad. Sci. USA 2008, 105, 2070 – 2075.
[6] W. Pao, V. A. Miller, K. A. Politi, G. J. Riely, R. Somwar, M. F.
Zakowski, M. G. Kris, H. Varmus, Plos Med. 2005, 2, 225 – 235.
[7] S. Kobayashi, T. J. Boggon, T. Dayaram, P. A. Jꢀnne, O. Kocher,
M. Meyerson, B. E. Johnson, M. J. Eck, D. G. Tenen, B. Halmos,
N. Engl. J. Med. 2005, 352, 786 – 792.
[8] G. Giaccone, Y. S. Wang, Cancer Treat. Rev. 2011, 37, 456 – 464.
[9] G. R. Oxnard, M. E. Arcila, J. Chmielecki, M. Ladanyi, V. A.
Miller, W. Pao, Clin. Cancer Res. 2011, 17, 5530 – 5537.
[10] L. V. Sequist, B. Besse, T. J. Lynch, V. A. Miller, K. K. Wong, B.
Gitlitz, K. Eaton, C. Zacharchuk, A. Freyman, C. Powell, R.
Ananthakrishnan, S. Quinn, J. C. Soria, J. Clin. Oncol. 2010, 28,
3076 – 3083.
8390 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 8387 –8390