Aminopyrimidinone Cdc7 Kinase Inhibitors

Article abstract of DOI:10.1016/j.bmcl.2012.01.041

We have investigated the SAR of a series of pyrimidinone-containing Cdc7 kinase inhibitors. A wide range of amine substitutions give potent compounds with activities (K_i) less than 1 nM. Kinase selectivity is reasonable and cytotoxicity corresponds to inhibition of MCM2 phosphorylation.

Full text of DOI:10.1016/j.bmcl.2012.01.041

Bioorganic & Medicinal Chemistry Letters 22 (2012) 1940–1943
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Aminopyrimidinone Cdc7 Kinase Inhibitors
⇑
Keith W. Woods , Chunqiu Lai, Julie M. Miyashiro, Yunsong Tong, Alan S. Florjancic,
Edward K. Han, Niru Soni, Yan Shi, Loren Lasko, Joel D. Leverson, Eric F. Johnson,
Alexander R. Shoemaker, Thomas D. Penning
Abbott Laboratories, 100 Abbott Park Rd., R47S AP10/307, Abbott Park, IL 60064-6101, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We have investigated the SAR of a series of pyrimidinone-containing Cdc7 kinase inhibitors. A wide range
of amine substitutions give potent compounds with activities (Ki) less than 1 nM. Kinase selectivity is rea-
sonable and cytotoxicity corresponds to inhibition of MCM2 phosphorylation.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 13 December 2011
Revised 12 January 2012
Accepted 13 January 2012
Available online 25 January 2012
Keywords:
Cdc7
Kinase
Pyrimidinone
Azaindole
Cancer
Cdc7 is a serine/threonine kinase that plays an essential role in
the initiation of DNA replication in eukaryotic cells. Cdc7, in con-
junction with the regulatory subunits Dbf4 or Drf1, is responsible
for phosphorylating and activating the putative helicase complex
MCM2-7. This step is required for the further assembly and func-
tional activation of the replication complex. Because tumor cells
often lack replication checkpoint mechanisms that would protect
normal cells, it has been hypothesized that Cdc7 inhibition would
bypass the replication checkpoint and selectively kill tumor cells.
For this reason, small molecule inhibitors of Cdc7 could be attrac-
tive candidates for therapeutic intervention in cancer and other
proliferative disorders.1–4
The SAR of the hit molecule 1 was explored with respect to each
ring, independently. Compounds with replacements for the phenyl
ring in compound 1 (ring A) were prepared as shown in Scheme 1
(see Table 1). Treatment of 4,6-dichloro-2-(methylthio)pyrimidine
(2) with sodium methoxide resulted in 4-chloro-6-methoxy-2-
(methylthio)pyrimidine (3). Heteroaryl replacements for the phe-
nyl ring were installed using Suzuki coupling conditions with the
appropriate heteroaryl boronic acid giving compounds 4. The sul-
fide (4) was oxidized to the sulfone (5) with OxoneÒ followed by
displacement with cyclohexyl amine to yield 8 (R = cyclohexyl,
R0 = H). O-demethylation gave the desired compounds 1, 9–14
(See Table 1).
There have been several reports of Cdc7 inhibitors in the past
few years.5–10 Our entry into the area began with the identification
of an aminopyrimidinone hit 1 (Cdc7 Ki = 230 nM) obtained from a
virtual screen of our compound library.11 SAR investigations were
carried out to optimize this structural series for Cdc7 activity and
selectivity. We now report a class of highly potent pyrimidinone-
containing Cdc7 inhibitors.
Investigation of the central ring modifications (ring B) was per-
formed by elaboration of five distinct nitrogen-containing six-
membered heteroaromatic rings as outlined in Scheme 2 (see Table
2, compounds 15–19, see supplementary data for more details).
Installation of the oxygen substitution was followed by amination
with cyclohexylamine and attachment of the 7-azaindole moiety
via palladium catalyzed Suzuki arylation conditions resulting in
compounds 15, 16, 18, and 19.12 The synthesis of compound 17
was more challenging. The first chloro displacement of 4-substi-
tuted 2,6-dichloropyridine with either alkoxide or cyclohexyl
amine proceeded smoothly, but the second chloro was resistant
to further displacement reactions. The desired compound was
ultimately obtained by the addition of t-BuOK to the 2-position
of 4-iodo-2,6-dichloropyridine followed by Suzuki reaction to
install the azaindole in the 4-position of the pyridine (17b). The
N-1 benzenesulfonyl protecting group on the azaindole was
O
NH
C
B
N
N
H
A
1
⇑
Corresponding author. Tel.: +1 847 938 2808; fax: +1 847 935 5165.
E-mail address: keith.w.woods@abbott.com (K.W. Woods).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2012.01.041

K. W. Woods et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1940–1943
1941
Table 2
Cl
OCH3
N
OCH3
Ring B (central ring) replacements
a
c
b
N
N
Ar
Cl
Ar
N
S
Cl
N
S
Ar
e
N
S
N
H
2
3
4
N
HN
Ar
OCH3
OCH3
N
O
Cdc7 Ki (nM)14
f
Example
N
N
NH
Ph
O
SO2CH3
Ar
N
Ph
Ar
N
NH
NH
14
<0.1
0.3
5
6
7
d
N
O
OCH3
N
O
N
N
15
e
NH
N
O
R
R
Ar
N
N
Ar
N
R'
R'
16
17
18
1.4
0.5
1.1
N
NH
NH
8
9-14, 20-29
O
O
Scheme 1. Reagents and conditions: (a) NaOMe/MeOH, rt; (b) aryl- or
heteroarylboronic acid, Pd(PPh3)4, K2CO3, DME/DMF, microwave, 130–170 °C; (c)
OxoneÒ, EtOAc/H2O; (d) NHRR0, dioxane, reflux; (e) 4 N HCl (aq), 80 °C; (f) PhMgBr,
DME, 0 °C.
HN
Table 1
Ring A (aromatic ring) replacements
OH
N
O
19
5.4
NH
Ar
Ar
N
N
H
Example
Cdc7 Ki (nM)14
in place of the cyclohexylamino group by allowing the sulfone 5
to react with phenyl magnesium bromide followed by O-demeth-
ylation (Scheme 1).
1
230
The compounds were found to bind competitively with ATP.
Although an X-ray structure of Cdc7 was not available, we did con-
sider a homology model throughout our SAR studies. Modeling
suggested that ring A was the kinase hinge-binding portion of
the inhibitors. Appropriately positioned nitrogen atoms are gener-
ally required for optimal interaction with what is known as the
hinge portion of the kinase. Compounds shown in Table 1 are con-
sistent with that hypothesis. Phenyl- (1) and pyridyl- (9) contain-
ing inhibitors lack a suitable nitrogen in a position that would be
able to form the desired hydrogen bonding and, as would be ex-
pected, show little activity. The heteroaromatic rings 4-pyridyl
(10), aminopyrimidinyl (11) and pyrazolyl (12) all gave modest
Cdc7 inhibitory activity. Overall, the 7-azaindole moiety (14) was
preferred.
Replacements for the B-ring offered some degree of flexibility
(Table 2). Even though we preferred the 2-amino-6-arylpyrimi-
din-4-one (14), most of the rings were tolerated. Even the least
active of the rings—2,6-disubstituted 4-hydroxypyridne (19)—
showed 5 nM activity. Modeling suggested that the central ring
oxygen plays a role in binding to the conserved Lys residue in
the Cdc7 binding pocket. A dramatic decrease in activity was ob-
served upon methylation of either the nitrogen or the oxygen of
the ring B pyrimidinone, in compounds like 14 (data not shown).
A broad range of amine substitutions (ring C) were tolerated. Al-
most any type of amine investigated was acceptable—aliphatic
amines, cycloalkyls, aromatics, and mono- or di-substituted amines
(Table 3). The only amine we found in this series to exhibit little
activity was the diphenylmethylamine (29, Ki = 330 nM). The
second phenyl ring apparently exceeds the spatial limitation of
the binding pocket.
9
>1200
2
N
10
N
N
N
11
5
NH2
12
13
15
N
N
N
H
0.5
N
H
N
14
<0.1
N
H
replaced with a SEM protecting group, and the cyclohexyl amine
was added, in low yield, using Buchwald amination conditions. Re-
moval of both the SEM and tert-butyl protecting groups yielded the
desired pyridinone 17.
Examples of replacements for the cyclohexylamino group (ring
C) are shown in Table 3 (compounds 20–29). The compounds were
prepared as described in Scheme 1. The ring A aromatic group in
each of these examples was 7-azaindole attached in the 3-posi-
tion.12 Example 7 was prepared with a direct phenyl attachment

1942
K. W. Woods et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1940–1943
A
X
Y
Cl
Z
N
H
15a: a
18a: b, c
19a: d, a
g, i
15b: X=Y=Z=N, A=OMe
18b: Y=Z=CH, X=N, A=OEt
19b: X=Y=CH, Z=N, A=OMe
OH
A
Z
X
Y
X
Y
Z
N
H
Cl
Cl
N
HN
16a: e, f, g
16b: a, i
17a: h, g
17b: j, k, m, n, p
15a: X=Y=Z=N, A=OMe
15: X=Y=Z=N
16a: X=Y=N, Z=CH, A=SMe
17a: X=Y=CH, Z=N, A=I
18a: X=Y=CH, Z=N, A= NH2
19a: X=Y=CH, Z=N, A=NO2
16: X=Y=N, Z=CH
17: X=Z=CH, Y=N
18: Y=Z=CH, X=N
19: X=Y=CH, Z=N
A
X
Y
Z
Cl
N
N
PhSO2
16b: X=Y=N, Z=CH, A=OMe
17b: X=Z=CH, Y=N, A=OtBu
Scheme 2. 13Reagents and reaction conditions: (a) cyclohexylamine, DIPEA, neat or DME, heat, microwave; (b) NaOEt/EtOH, sealed vessel, 150 °C; (c) cyclohexanone, HOAc,
Na(OAc)3BH, THF; (d) MeOH, K2CO3, heat; (e) mCPBA, CH2Cl2; (f) NaOMe/MeOH, ꢀ78 °C; (g) 1-benzenesulfonyl-7-azaindole-3-boronic acid pinacol ester,12 PdCl2(dppf) or
(PPh3)2Pd(II)Cl2, DME/EtOH/H2O, microwave, 130–170 °C; (h) KOtBu, THF, reflux; (i) 4 N HCl (aq), heat; (j) KOH, EtOH/H2O, 50 °C; (k) SEM-Cl, NaH, DMF, 0 °C–rt; (m)
Pd2(dba)3, Xantphos, KOtBu, tBuOH, microwave, 130 °C; (n) TBAF, ethylenediamine, DMF, 85 °C; (p) TFA/CH2Cl2, rt.
Table 3
Ring C (amine) replacements
O
NH
N
R
N
HN
Example
R
Cdc7 Ki (nM)14
cell viability EC50
0.103
(l
M)15
pMCM2 EC50
0.06
(l
M)16
14
<0.07
HN
20
0.09
0.106
0.015
N
CH3
21
22
23
24
25
0.12
<0.07
0.18
0.443
0.122
2.21
0.143
0.18
0.06
0.30
0.30
HN
HN
N
O
O
O
0.18
0.45
N
O
0.14
2.46
HN
HN
N
26
0.11
7.37
2.61
7
0.25
0.2
0.83
6.19
0.41
1.78
OH
27
N

K. W. Woods et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1940–1943
1943
Table 3 (continued)
Example
R
Cdc7 Ki (nM)14
cell viability EC50
0.52
(l
M)15
pMCM2 EC50
0.30
(l
M)16
O
CH3
28
29
0.2
N
Ph
Ph
HN
328
ND
ND
6. Shafer, C. M.; Lindvall, M.; Bellamacina, C.; Gesner, T. G.; Yabannavar, A.; Jia,
W.; Lin, S.; Walter, A. Bioorg. Med. Chem. Lett. 2008, 18, 4482.
Select Cdc7 inhibitors were additionally assayed in a cell viabil-
ity assay using HCT116 cells to assess the cytotoxicity of the com-
pounds.15 Compounds were additionally assayed for their ability to
inhibit MCM2 phosphorylation.16 MCM2 is one of the cellular tar-
gets phosphorylated by Cdc7/Dbf4. Cytotoxic compounds acting
through a Cdc7 mediated process should show inhibition of phos-
phorylation of MCM2. Those data are shown in Table 3. General cell
killing with little affect on MCM2 is likely to be due to off-target
activity.
7. Menichincheri, M.; Bargiotti, A.; Berthelsen, J.; Bertrand, J. A.; Bossi, R.;
Ciavolella, A.; Cirla, A.; Cristiani, C.; Croci, V.; D’Alessio, R.; Fasolini, M.;
Fiorentini, F.; Forte, B.; Isacchi, A.; Martina, K.; Molinari, A.; Montagnoli, A.;
Orsini, P.; Orzi, F.; Pesenti, E.; Pezzetta, D.; Pillan, A.; Poggesi, I.; Roletto, F.;
Scolaro, A.; Tatò, M.; Tibolla, M.; Valsasina, B.; Varasi, M.; Volpi, D.;
Santocanale, C.; Vanotti, E. J. Med. Chem. 2009, 52, 293.
8. Ermoli, A.; Bargiotti, A.; Brasca, M. G.; Ciavolella, A.; Colombo, N.; Fachin, G.;
Isacchi, A.; Menichincheri, M.; Molinari, A.; Montagnoli, A.; Pollan, A.; Rainoldi,
S.; Sirtori, F. R.; Sola, F.; Thieffine, S.; Tibolla, M.; Valsasin, B.; Volpi, D.;
Santocanale, C.; Vanotti, E. J. Med. Chem. 2009, 52, 4380.
Since most of the compounds we examined were potent it was
useful to prioritize compounds with respect to kinase selectivity.
Compound 14 (Ki <0.07 nM), was the most potent compound in
this series but was found to be unselective when tested against a
panel of around 125 kinases. Compound 14 had activity (Ki) of less
than 10 nM against 52 kinases out of a panel of more than 125 ki-
nases. We were able to achieve good improvements in selectivity
in several cases. Replacing the cyclohexylamine (14) with 4-
hydroxypiperidine (27) or 4-methoxypiperidine (28) resulted in
inhibitors that had activity (Ki) of less than 10 nM against only 7
out of 125 kinases tested and bismethoxyethylamine (23) showed
activity (Ki) against only 9 kinases in the same range of activity.
Pharmacokinetic profiles were obtained for two of the most
selective compounds (23 and 28). Those compounds were both
characterized by short half-lives and no oral bioavailability. Part
of the reason for the short half-lives could be attributed to appar-
ent glucuronidation. The poor PK properties limited our ability to
obtain meaningful results for this series of compounds in in vivo
efficacy trials.
We have reported a series of pyrimidinone-containing ATP
competitive Cdc7 inhibitors. A wide range of amines give com-
pounds with activities (Ki) less than 1 nM. We have been able to
obtain reasonable kinase selectivity and cytotoxicity generally cor-
responds to inhibition of MCM2 phosphorylation. Poor PK proper-
ties have limited our ability to study the efficacy of these inhibitors
in vivo. Additional efforts to address deficiencies in both selectivity
and PK will be reported in subsequent publications.
9. Zhao, C.; Tovar, C.; Yin, X.; Xu, Q.; Todorov, I. T.; Vassilev, L. T.; Chen, L. Bioorg.
Med. Chem. Lett. 2009, 19, 319.
10. Menichincheri, M.; Albanese, C.; Alli, C.; Ballinari, D.; Bargiotti, A.; Caldarelli,
M.; Ciavolella, A.; Cirla, A.; Colombo, M.; Colotta, F.; Croci, V.; D’Alessio, R.;
D’Anello, M.; Ermoli, A.; Fiorentini, F.; Forte, B.; Galvani, A.; Giordano, P.;
Isacchi, A.; Martina, K.; Molinari, A.; Moll, J. K.; Montagnoli, A.; Orsini, P.; Orzi,
F.; Pesenti, E.; Pillan, A.; Roletto, F.; Scolaro, A.; Tatò, M.; Tibolla, M.; Valsasina,
B.; Varasi, M.; Vianello, P.; Volpi, D.; Santocanale, C.; Vanotti, E. J. Med. Chem.
2010, 53, 7296.
11. Muchmore, S. W.; Debe, D. A.; Metz, J. T.; Brown, S. P.; Martin, Y. C.; Hajduk, P. J.
J. Chem. Inf. Model. 2008, 48, 941.
12. Azaindole was brominated in the 3-position with NBS in THF. The material was
then allowed to react with benzenesulfonylchloride in DMF under basic (KOH)
conditions to protect the N-1 nitrogen. Treatment with bis(pinacolato)diboron,
(PPh3)2Pd(II)Cl2, and KOAc in THF yielded the azaindole pinacolborate used for
Suzuki coupling reactions.
13. See Supplementary data for more detailed synthetic schemes for compounds
15–19.
14. Cdc7 Activity Assay: Cdc7 (BEV coexpressed huCDC7/DBF4) was prepared
internally. In 384-well v-bottom polypropylene plates 6
2% DMSO) was mixed with 6 L of Cdc7 (2 g/mL) and Jerini peptide substrate
A-A11 (biotin-C6linker-TPSDSLIYDDGLS) (Final 2 M) followed by immediate
initiation with 6
L -[33P]-ATP (Final 1 M, 20 mCi/mol) using a reaction buffer
comprising 20 mM HEPES, pH 7.5, 1 mM DTT, 10 mM MgCl2, 100 M Na3VO4,
0.075% Triton X-100. Reactions were quenched after 1 h by the addition of
90
lL compound (Final
l
l
l
l
l
L stop buffer (33.3 mM EDTA, 1.33 M NaCl in 1ꢁ PBS buffer). 85 L of the
stopped reactions were transferred to 384-well streptavidin-coated plates
(FlashPlate Plus, Perkin–Elmer), incubated 10 min at room temperature and
washed 3 times with 0.05% Tween-20/PBS using an ELX-405 automated plate
washer (BioTek), and counted on
a TopCount Scintillation Plate Reader
(Packard). IC50 values were determined via non-linear regression fitting of
enzyme inhibition data and corresponding Ki values were generated assuming
ATP-competitive (equilibrium) inhibition and using the experimentally
determined apparent ATP Km of 0.7
15. Cell Proliferation Assay: HCT-116 cells were plated in a 96-well plate at 3000
cells (100 L) per well and placed in an incubator for 24 h at 37 °C. The next
lM.
l
day, cells were treated with serially diluted concentrations of Cdc7 inhibitors.
After 3 days, cell viability was determined by adding an equal volume of Cell
Titer GLO (Promega, Madison, WI). The plate was read on a SpectraMax M5
(Molecular Devices, Sunnyvale, CA). The EC50 for each compound was
determined based on the viability readout.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2012.01.041.
16. pMCM2 Assay Protocol: HCT-116 cells were plated in a 96-well plate at 20,000
cells/well. The next day, cells were treated with inhibitors for 6 h at 37 °C.
Medium was removed and cells were lysed with 20
l
L
of lysis buffer
of lysed
References and notes
(Biosource), followed by 80 of PBS containing 2% BSA. 80 lL
lL
sample was transferred to an anti-MCM2 coated plate. Anti-MCM2 coated
plates were generated by coating Reacti-bind plates (Thermo Scientific,
Rockford, IL) with anti-MCM2 antibody (Bethyl Laboratories, Montgomery,
TX) (1:100), followed by washing (PBS containing 0.05% Tween 20) and
blocked with blocking buffer (2% BSA in PBS). After addition of lysed sample,
1. Sawa, M.; Masai, H. Drug Design Dev. Ther. 2008, 2, 255.
2. Kim, J. M.; Yamada, M.; Masai, H. Mutation Res. 2003, 532, 29.
3. Ito, S.; Taniyama, C.; Arai, N.; Masai, H. Drug News Perspect. 2008, 21, 481.
4. Montagnoli, A.; Moll, J.; Colotta, F. Clin. Cancer Res. 2010, 16, 4503.
5. Vanotti, E.; Amici, R.; Bargiotti, A.; Berthelsen, J.; Bosotti, R.; Ciavolella, A.; Cirla,
A.; Cristiani, C.; D’Alessio, R.; Forte, B.; Isacchi, A.; Martina, K.; Menichincheri,
M.; Molinari, A.; Montagnoli, A.; Orsini, P.; Pillan, A.; Roletto, F.; Scolaro, A.;
Tibolla, M.; Valsasina, B.; Varasi, M.; Volpi, D.; Santocanale, C. J. Med. Chem.
2008, 51, 487.
the plate was incubated for 16 h at 4 °C, washed, and then 100 lL of anti-
pMCM2 antibody was added. The plate was incubated for 2 h at room
temperature, washed, and goat anti-rabbit HRP added. The plate was read on a
SpectraMax M5 (Molecular Devices, Sunnyvale, CA).