A novel approach for the development of selective Cdk4 inhibitors: Library design based on locations of Cdk4 specific amino acid residues

Article abstract of DOI:10.1021/jm010326y

Identification of a selective inhibitor for a particular protein kinase without inhibition of other kinases is critical for use as a biological tool or drug. However, this is very difficult because there are hundreds of homologous kinases and their kinase

Full text of DOI:10.1021/jm010326y

4628
J . Med. Chem. 2001, 44, 4628-4640
A Novel Ap p r oa ch for th e Develop m en t of Selective Cd k 4 In h ibitor s: Libr a r y
Design Ba sed on Loca tion s of Cd k 4 Sp ecific Am in o Acid Resid u es
Teruki Honma,* Takashi Yoshizumi, Noriaki Hashimoto, Kyoko Hayashi, Nobuhiko Kawanishi,
Kazuhiro Fukasawa, Tohru Takaki, Chinatsu Ikeura, Mari Ikuta, Ikuko Suzuki-Takahashi, Takashi Hayama,
Susumu Nishimura, and Hajime Morishima
Banyu Tsukuba Research Institute in collaboration with Merck Research Laboratories, Okubo-3,
Tsukuba 300-2611, Ibaraki, J apan
Received J uly 17, 2001
Identification of a selective inhibitor for a particular protein kinase without inhibition of other
kinases is critical for use as a biological tool or drug. However, this is very difficult because
there are hundreds of homologous kinases and their kinase domains including the ATP binding
pocket have a common folding pattern. To address this issue, we applied the following structure-
based approach for designing selective Cdk4 inhibitors: (1) identification of specifically altered
amino acid residues around the ATP binding pocket in Cdk4 by comparison of 390 representative
kinases, (2) prediction of appropriate positions to introduce substituents in lead compounds
based on the locations of the altered amino acid residues and the binding modes of lead
compounds, and (3) library design to interact with the altered amino acid residues supported
by de novo design programs. Accordingly, Asp99, Thr102, and Gln98 of Cdk4, which are located
in the p16 binding region, were selected as first target residues for specific interactions with
Cdk4. Subsequently, the 5-position of the pyrazole ring in the pyrazol-3-ylurea class of lead
compound (2a ) was predicted to be a suitable position to introduce substituents. We then
designed a chemical library of pyrazol-3-ylurea substituted with alkylaminomethyl groups based
on the output structures of de novo design programs. Thus we identified a highly selective and
potent Cdk4 inhibitor, 15b, substituted with a 5-chloroindan-2-ylaminomethyl group. Compound
15b showed higher selectivity on Cdk4 over those on not only Cdk1/2 (780-fold/190-fold) but
also many other kinases (>430-fold) that have been tested thus far. The structural basis for
Cdk4 selective inhibition by 15b was analyzed by combining molecular modeling and the X-ray
analysis of the Cdk4 mimic Cdk2-inhibitor complex. The results suggest that the hydrogen
bond with the carboxyl group of Asp99 and hydrophobic van der Waals contact with the side
chains of Thr102 and Gln98 are important. Compound 15b was found to cause cell cycle arrest
of the Rb(+) cancer cell line in the G1 phase, indicating that it is a good biological tool.
In tr od u ction
dicted binding mode. The predicted binding mode was
validated by the X-ray analysis of the Cdk2-1 complex
(PDB ID: 1GIH)9 and the structure-activity relation-
ship (SAR) (Figure 1).
Cyclins and cyclin-dependent kinases (Cdks) play
important roles in regulation of the cell cycle.1 In
particular, D-type cyclins, which have been shown to
be amplified or overexpressed in several tumor cells,2
associate with Cdk4/6 to activate their phosphorylation
activity. Cyclin D-Cdk4/6 complexes phosphorylate the
retinoblastoma protein (pRB) and regulate the cell cycle
during G1/S transition.3 Loss of function or deletion of
p16ink4a (endogenous Cdk4/6 specific inhibitor protein)
frequently occurs in clinical cancer cells.4 Thus, selective
Cdk4/6 inhibitors should be useful as a new class of
cytostatic antitumor agents.5
In the previous paper, we reported the structure-
based lead generation of a novel diarylurea class of
potent Cdk4 inhibitors.6 The Cdk4 homology model was
constructed according to X-ray analysis of the activated
form of Cdk2.7 Using this model, we applied a new de
novo design strategy8 to generate new scaffold candi-
dates. As a result, a diarylurea class of scaffold was
identified, and a potent compound 1 (IC50 ) 0.042 µM)
was obtained through modifications based on the pre-
Compound 1 showed good Cdk4 inhibitory activity
and moderate selectivity (>50-fold) over other kinases
except the Cdk family (Table 1). However, 1 also
inhibited Cdk family kinases in addition to Cdk4/6,
namely Cdk1 and Cdk2, with comparable IC50 values.
In the predicted binding mode, compound 1 appeared
to form hydrogen bonds with the main chain of Val96
and the side chain amino group of Lys35. Although
Val96 in Cdk4 is replaced by Leu in Cdk1/2, the main
chain atoms are located in the same position. Lys35 is
conserved in Cdk4 and Cdk1/2. Additionally, since the
sequence identities between Cdk4 and Cdk1/2 are high
(43%/45%), the shapes of their ATP binding pockets
were predicted to be very similar. Therefore, it was
assumed that compound 1 does not achieve any specific
interactions with Cdk4 against Cdk1/2. The selectivity
of Cdk4(/6) over Cdk1/2 is thought to be critical for the
arrest of the cell cycle in the G1 phase because Cdk1
and Cdk2 regulate the transition to other phases of the
cell cycle.10 To suppress tumor growth, G1 arrest is
expected to reduce the stress of normal cells other than
* To whom correspondence should be addressed. Tel: 81-298-77-
2000. Fax: 81-298-77-2029. E-mail: honmatr@banyu.co.jp.
10.1021/jm010326y CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/13/2001

Development of Selective Cdk4 Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26 4629
arrest in other phases because normal cells are usually
resting in the G0-G1 phase.11 In this report, we tried to
design Cdk4 selective inhibitors with effects over both
Cdk1/2 and other kinases.
Enhancement of selectivity for the target protein over
its highly homologous counterparts is not feasible
because there are few differences in the protein surface
of the ligand binding sites. Recently, successful ex-
amples of the development of such selective compounds
have been reported; these processes used structural
information about critical differences in amino acid
residues and/or conformations between the target pro-
teins (MMP-13, PTP1B, and COX-2) and their counter-
parts.12 However, in the case of protein kinases, it is
more difficult to improve selectivity for a target protein
because there are hundreds of homologous kinases in
the superfamily.13 To overcome this problem, we initially
identified specific amino acid residues around the ATP
binding pocket of Cdk4 by comparing the amino acid
sequences of 390 representative kinases. Subsequently,
a chemical library was designed using this information
about the locations of these amino acid residues and the
binding mode of compound 1 to achieve specific interac-
tions with Cdk4, discriminating against interactions
with other kinases.
Resu lts a n d Discu ssion
Id en t ifica t ion of Sp ecifica lly Alt er ed Am in o
Acid Resid u es a r ou n d th e ATP Bin d in g P ock et of
Cd k 4. First, we performed a 3D superposition of the
structural model of Cdk4 bound with compound 1 and
the X-ray structure of Thr160 phosphorylated Cdk2-
cyclin A-ATPγS reported by A. A. Russo et al.7 We then
chose 39 amino acid residues in the Cdk4 model which
are located within 7 Å of the ATPγS molecule in the
Cdk2 (these residues are shown in Figure 2).
F igu r e 1. Structure and binding mode of compound 1. A:
structure of compound 1. B: the binding mode of 1 in the Cdk4
homology model based on X-ray analysis of the Cdk2-1
complex (PDB ID: 1GIH).9 Green residues: residues conserved
in Cdk4 and Cdk1/2. Red residues: residues altered in Cdk1/
2. Broken lines: predicted hydrogen bonds.
Ta ble 1. Potency and Cdk4 Selectivity of Compound 1 for Representative Kinases
a
These data are based on racemic 1.

4630 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26
Honma et al.
F igu r e 2. Amino acid residues around the ATP binding pocket. Numbers in rows of AGC-OPK groups and 390 kinases show
conserved ratios. Red cells: residues that are replaced in Cdk1/2/6 or <40% conserved in 390 kinases. Box: residues in contact
with or accessible to ATPγS or compound 1. Yellow circles: residues that are <40% conserved and are in contact with or are
accessible to either ATPγS or compound 1. Red circles: residues that are replaced in both Cdk1 and Cdk2 and are in contact with
or accessible to either ATPγS or compound 1. Red open circle: residues that are replaced in Cdk1 and are in contact with or
accessible to either ATPγS or compound 1. aSequence identity confined to around the ATP binding pocket of Cdk4. bSequence
identity based on full sequences.
The multiple alignment of 390 representative kinase
sequences by S. K. Hanks and A. M. Quinn was pub-
lished on the Internet.14 Using the multiple alignment,
we calculated (1) the conserved ratios for each of the
39 residues of Cdk4 in the 390 kinases and (2) the amino
acid identities in the region confined to the 39 residues
around the ATP binding pocket. The multiple alignment
among Cdk4, Cdk6, Cdk1, and Cdk2 and the results of
the calculations are summarized in Figure 2. From these
data, we were able to identify frequently altered amino
acid residues in the 390 kinases and altered amino acid
residues in Cdk1/2. Among the frequently altered
residues, those with side chains not facing inhibitors
are not useful for the design of new inhibitors. We must
use the differences of chemical properties and/or shape
of the residues' side chains that faced compound 1 to
improve Cdk4 selectivity. For this purpose, 26 residues
that were directly in contact with or accessible to ATPγS
or compound 1 were selected by considering the 3D
structures of the complexes. These residues are enclosed
in the boxes in Figure 2. Thirteen residues (Ile12, Val14,
Ala16, Tyr17, Phe93, His95, Val96, Asp97, Gln98,
Asp99, Thr102, Glu144, and Ala157), which are <40%
conserved in 390 kinases and in contact with or acces-
sible to ATPγS or compound 1, are shown by yellow
circles in Figure 2. Among the 13 residues, Ile12, Tyr17,
Phe93, and Asp99 are highly conserved in the CMGC
group (Hanks’ classification of kinases13), which includes
Cdk4. Tyr17 is thought to be involved in regulation of
phosphorylation activity of the Cdks and related ki-
nases.15 The side chains of Phe93, His95, Val96, and
Ala157 were already predicted to have contact with
compound 1 from X-ray analysis of the Cdk2-1 complex.
The moderately higher selectivity of compound 1 for
Cdk4 over other kinases is likely to be due to specific
interactions with these residues.
Subsequently, we tried to identify altered amino acid
residues in Cdk4 and Cdk1/2. In the region of the ATP
binding pocket of Cdk4, residue identities were found
to be very high with respect to Cdk1/2 (64%/72%), while
the identities throughout the full sequence were not as
high (43%/45%). Among the residues around the ATP
binding pocket, we identified eight residues (Val14,
Ala16, His95, Val96, Asp97, Gln98, Thr102, and Glu144)
which are altered in Cdk1/2 and have contact or are
accessible to ATPγS or compound 1 (these residues are
shown by red circles and a red open circle in Figure 2).
Except for Gln98, all of these residues are altered in
both Cdk1 and Cdk2. Interestingly, Gln98 is conserved
in Cdk2 as shown in Figure 3C. It is noteworthy that
five of eight residues are distributed on one side of the
ATP binding pocket. According to X-ray analyses of
Cdk6-p16/p19 complexes, this side is the p16 binding
region (Figure 3B).16 p16 is a Cdk4/6-specific, endog-
enous inhibitor protein that is thought to suppress
carcinogenesis.17 Therefore, the amino acid residues in
the p16 binding region of Cdk4 are likely to interact
with p16 in a specific way.

Development of Selective Cdk4 Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26 4631
F igu r e 3. Locations of altered residues and the binding mode of 1. A: locations of altered/frequently altered residues around the
ATP binding pocket of Cdk4. Yellow residues: residues that are <40% conserved in 390 kinases and are in contact with or accessible
to either ATPγS or compound 1. Red residues: residues that are replaced in Cdk1 or Cdk2, <40% conserved in 390 kinases, and
are in contact with or accessible to either ATPγS or compound 1. B: superimposed model of the Cdk4-ATPγS-p16 complex.
Green ribbon: Cdk4 model. Cyan ribbon: p16. CPK representation: ATPγS. The location of ATPγS and p16 were predicted by
superposition of the Cdk4 homology model, Thr160 phosphorylated Cdk2-cyclinA-ATPγS (PDB ID: 1J ST), and Cdk6-p16 (PDB
ID: 1BI7). C: the binding mode of compound 1 in the Cdk4 homology model and altered/frequently altered residues in the p16
binding region. The Cdk4 homology model: solvent accessible surface (probe: 1.4 Å) representation (colored by partial charge).
Compound 1: stick representation.
By analyzing the binding mode of compound 1, we
found that the pyridine ring of compound 1 is directed
toward the amino acid residues in the p16 binding
region (Figure 3C). The terminal of the side chain of
Val96 cannot come in contact with the inhibitor because
it is hidden by the side chain of Leu147. Ile12 is replaced
by Leu in most kinases, including Cdk1/2; this causes
only small changes in the protein surface. For these
reasons, we removed Val96 and Ile12 from target
residues. On the other hand, Asp97 is replaced by His
(Cdk2) or Ser (Cdk1), and Thr102 is replaced by Lys
(Cdk1/2). These alterations cause dramatic changes of
the chemical properties and shape of the protein surface.
Amino acid residues such as Asp99, Gln98, and His95
are also useful. Asp99 is replaced by Ala, Ser, Asn, Pro,
or Cys in most tyrosine kinases. Gln98 is replaced by
Met in Cdk1 and by various other residues in other
kinases. His95 is replaced by Phe in Cdk1/2 and by Phe
or Tyr in most kinases. In this regard, in order to
enhance Cdk4 selectivity against Cdk1/2 and other
kinases, the introduction of substituents that face the
side chains of these residues was expected to be effec-
tive.
Libr a r y Design Ba sed on Str u ctu r a l In for m a -
tion . To select first target residues to interact with
substituents on our lead compounds, we calculated the
shortest distances between the substitutable positions
of the inhibitors and the side chain of each residue in
the p16 binding region (Figure 4B). We found that
compounds 2a and 2b, which have a five-membered
pyrazole ring, are useful as leads in addition to com-
pound 1 in order to approach these residues from
various directions. The structures and inhibitory activi-
ties for Cdk4 of 2a and 2b are shown in Figure 4A, and
the binding mode of 2a predicted by a docking study is
shown in Figure 4B. Asp99, Thr102, and Gln98 were
more accessible than the other residues. On the basis
of the binding modes, the 5-position of the pyrazole ring

4632 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26
Honma et al.
F igu r e 4. Predicted suitable substitution positions and substituents. A: structures of 2a and 2b and their IC50 (Cdk4) values.
B: predicted binding modes of compound 1/2a and suitable positions to introduce substituents. Green stick: compound 1. Blue
stick: compound 2a . Left numbers: shortest distances between substitutable positions of the pyridine ring in compound 1 and
the side chain of each residue. Right numbers: shortest distances between substitutable positions of the pyrazole ring in compound
2a and the side chain of each residue. Arrows: directions of substituents on the pyridine ring and pyrazole ring. Bold arrow:
predicted most appropriate direction toward Asp99, Thr102, and Glu98. C: schematic figure of binding mode of compound 2a
and appropriate substituent candidates. Bold line: a typical output structure interacted with Asp99, Thr102, and Glu98 that
was proposed by de novo design programs (LUDI and LeapFrog).
in compound 2a (blue bold arrow in Figure 4B) was
predicted to be the most appropriate position and
direction to contact with the side chains of the three
residues. Therefore, we constructed a 5-substituted
pyrazol-3-yl urea library to enhance Cdk4 selectivity.
reductive amination with the corresponding aldehyde
of compound 2a . Thus we constructed a chemical library
of 2a featuring a pyrazole ring with alkylaminomethyl
substituents at the 5-position.
The synthetic routes of alkylaminomethyl library 9
and the cyclized amino substituted compound 14 are
summarized in Scheme 1. Compound 4 was prepared
by the sequential coupling of 3, acetonitrile, and hydra-
zine.21 Protection of 4 by the Boc group followed by a
coupling reaction with 9-amino-1,2,3,9b-tetrahydro-5H-
pyrrolo[2,1-a]isoindol-5-one6 provided the diarylurea
compound 6 as a racemate. Deprotection and oxidation
by MnO2 gave aldehyde 8, which was a substrate for
reductive amination. The imine formation of 8 by
treatment with alkylamines and MS3Å followed by a
reduction provided the alkylamino library 9. Synthesis
of the cyclized amino substituted compound 14 was
initiated from the optically active (S)-proline derivative.
The diarylurea compound 13 was prepared by a proce-
dure similar to that for synthesis of 6. Deprotection of
13 gave compound 14.
SAR P r ofiles a n d Str u ctu r a l Ba sis for Selective
Cd k 4 In h ibition . Compounds in the first alkylamino
library (64 compounds) were tested by both cyclin
D-Cdk4 and cyclin A-Cdk2 inhibitory assay. The
potencies and Cdk4 selectivities over Cdk2 of the
representative compounds are summarized in Table 2.
Interestingly, introduction of the amino groups caused
To design suitable substituents, we employed the de
novo design programs,18 LUDI19 and LeapFrog.20 Al-
though output structures from de novo design programs
are often not synthetically feasible, these structures
suggest suitable properties such as a hydrogen-bonding
donor, acceptor, and hydrophobicity and appropriate
lengths of substituents. The aminomethyl groups and
their cyclized groups were provided as typical structures
by both LUDI and LeapFrog (Figure 4C). The NHs of
these amino groups were predicted to interact with the
carboxyl group of Asp99 and/or the side chain OH of
Thr102. The OH of Thr usually functions as both a
hydrogen-bonding donor and acceptor. However, in this
case, the OH of Thr102 was already predicted to make
a hydrogen bond with the carboxyl group of Asp99, and
the side chain of Thr102 was expected to work as a
hydrogen-bonding acceptor and hydrophobic portion for
our lead compounds. As alkyl groups via the amino-
methyl group, hydrophobic substituents were thought
to be appropriate considering the hydrophobic environ-
ment surrounded by the side chain methyl group of
Thr102 and the side chain methylene group of Gln98.
These alkylamino groups can be easily introduced by

Development of Selective Cdk4 Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26 4633
Sch em e 1. Synthetic Routes of 5-Substituted Pyrazole-3-yl Urea Derivativesa
a
Reagents: (a) (1) BuLi, CH3CN, tetrahydrofuran; (2) NH2NH2, EtOH, 40% (for 4), 81% (for 11); (b) NaH, (Boc)2O, tetrahydrofuran,
53% (for 5), 61% (for 12); (c) (1) 4-(dimethylamino)pyridine, p-nitrophenyl chloroformate, CHCl3; (2) amine;25 (3) 4 N HCl, MeOH, 19%;
(d) Pd(OH)2/C, H2, 2 N HCl, MeOH-tetrahydrofuran, 93%; (e) MnO2, CHCl3-dimethylformamide, 59%; (f) (1) alkylamine, MS3Å,
chloroform-MeOH; (2) NaBH4, 20-50%; (g) (1) NaH, 4-nitrophenyl chloroformate, tetrahydrofuran; (2) amine,25 4-(dimethylamino)pyridine,
42%; (3) 6 N HCl, MeOH, 21%; (h) Pd/C, H2, 4 N HCl, MeOH, 51%.
Ta ble 2. IC50 Values and Cdk4 Selectivity over Cdk2 of
5-Alkylaminomethyl-pyrazole-3-yl Urea Derivatives
caused reduction of Cdk4 inhibitory activity in general,
the compounds substituted with tert-butyl (9d ) and
cyclopentyl (9e) had almost comparable inhibition po-
tencies to that of compound 1. Consequently, 9d and
9e showed better Cdk4 selectivities over Cdk2 (180-fold
and 150-fold, respectively). These SARs and structural
information around the p16 binding region suggested
that (1) NH2+s on the alkylamino groups formed a
hydrogen bond with the carboxyl group of Asp99 (Asp86
in Cdk2), (2) in Cdk2, the alkylamino groups con-
strained by the hydrogen bond with Asp86, caused
electrostatic and/or steric repulsion with the side
chain of Lys89, and (3) in Cdk4, the hydrophobic and
bulky substituents of 9d and 9e fit into the hydrophobic
region surrounded by the side chains of Thr102 and
Gln98.
From these facts, we designed 14a and 14b with a
pyrrolidine ring to restrict the substituent, and 15a and
15b with a 5-chloroindan-2-ylaminomethyl group to
enhance the hydrophobic interaction with Cdk4, while
maintaining selectivity over Cdk2. Each compound was
prepared as a single diastereomer22 by chiral HPLC
separation. The potency and Cdk4 selectivity over
Cdk1/2 of each compound are summarized in Table 3.
Compound 14a and 14b showed comparable selectivities
over Cdk2 (110-fold/120-fold) to that of 9d and 9e. We
successfully completed X-ray analysis of the mutant
Cdk2 (Cdk4 mimic Cdk2, having F82H, L83V, and
K89T) -14b complex (PDB ID: 1GIJ ) using the soaking
+
method (Figure 5A).9 In this X-ray structure, the NH2
group of pyrrolidine in 14b formed a salt bridge with
the carboxyl group of Asp86, which proved our hypoth-
esis. On the other hand, 15a /15b showed 30-40-fold
enhancement of Cdk4 inhibitory activity (0.0016 µM/
0.0023 µM) compared to that of 9e with good selectivity
over Cdk2. The docking studies23 between Cdk4/Cdk2
and 15b predicted steric repulsion between the large
chloroindanyl group constrained by the hydrogen bond
a 100-200-fold decrease in Cdk2 inhibitory activity as
compared to that of compound 1 (IC50 (Cdk2) ) 0.078
µM). Although addition of these amino groups also

4634 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26
Honma et al.
Ta ble 3. IC50 Values and Cdk4 Selectivity over Cdk1/2 of Designed Compounds
with Asp86 and the side chain of Lys89 in Cdk2 (Figure
5C). In addition, it was predicted that the chloroindanyl
group in 15b will make good van der Waals contact with
the side chains of Thr102 and Gln98 in the Cdk4 model.
This contact between hydrophobic surfaces possibly
contributes to the enhanced binding affinity of 15b with
Cdk4. Interestingly, 15b showed higher selectivity over
Cdk1 (780-fold) than over Cdk2 (190-fold), while 14a
showed almost equal selectivity over both Cdk1 (130-
fold) and Cdk2 (120-fold). As already shown in Figure
3C, Gln98 in Cdk4 is replaced by Met in Cdk1 but is
conserved in Cdk2. This difference between Cdk1 and
Cdk2 may cause different inhibitory activities between
Cdk1 (0.44 µM) and Cdk2 (1.8 µM).
Con clu sion
We developed a highly selective and potent Cdk4
inhibitor 15b by structure-based drug design. Com-
pound 15b showed excellent selectivity not only over
Cdk1/2 (780-fold/190-fold) but also over many other
kinases (>430-fold) so far investigated. Cell-based stud-
ies revealed that 15b of between 0.1 and 0.5 µM
concentration causes G1 arrest in the Rb(+) cancer cell
line (Molt-4). Further details regarding the cellular
profiles of these Cdk4 selective inhibitors will be re-
ported in a following manuscript.23
Identification of altered residues in Cdk1/2 and in 390
other kinases in their ATP binding pockets allowed us
to focus on target residues to interact with substituents
on lead compounds. Library design based on the loca-
tions of these residues and the binding modes of lead
compounds enabled us to develop potent and selective
compounds efficiently.
The kinase superfamily consists of numerous kinases
that have a common folding pattern. Therefore, in
general it is difficult to improve target kinase selectivity
with respect to all the other kinases at the same time.
Our approach should be helpful for the development of
specific kinase inhibitors more systematically and ef-
ficiently.
Finally, we investigated the Cdk4 selectivity of 15b
over other representative kinases. Schematic figures of
the binding modes of the key compounds 1 and 15b,
their Cdk4 selectivity over representative kinases, and
the sequence alignment (Asp99, Thr102, and Gln98) are
summarized in Figure 6. The selectivity of 15b over
Cdk1/2 was remarkably higher than that of compound
1, while the selectivity of 15b over other Ser/Thr kinases
and Tyr kinases remained the same or were enhanced.
+
The NH2 of 15b, which is assumed to interact with
Asp99, is critical for the higher Cdk4 selectivity not only
over Cdk1/2 but also over other kinases. As shown in
Figure 6, this Asp99 is replaced by the nonacidic
residues in most tyrosine kinases. Thr102 and Gln98
are also frequently replaced by other residues in both
Ser/Thr kinases and tyrosine kinases. The interactions
between the three residues and 15b were supposed
to contribute to the improvement in Cdk4 selectivity
over Lck (3000-fold), Flt-1 (480-fold), FGFR1 (7000-fold),
and PDGFRâ (1500-fold) compared with those of com-
pound 1.
Exp er im en ta l Section
Ma ter ia ls a n d Meth od s. All reagents and solvents were
of commercial quality and were used without further purifica-
tion unless otherwise noted. Melting points were determined
with a Yanaco MP micromelting point apparatus (Yanagimoto
Seisakusho Co., Ltd., J apan) and were not corrected. 1H NMR
spectra were obtained on a Varian Gemini-300 (300 MHz) or
Gemini-200 (200 MHz) with tetramethylsilane as an internal
standard. Mass spectrometry was performed with a J EOL
J MS-SX 102A (FAB positive) or micromass QUATTRO II (ESI
positive). High-resolution mass spectra (HR-MS) were deter-

Development of Selective Cdk4 Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26 4635
F igu r e 5. Structural basis of selective Cdk4 inhibition by compound 15b. A: binding mode of 14b in Cdk4 mimic Cdk2 (X-ray,
9
PDB ID: 1GIJ ). B: binding mode of 15b in Cdk4 (model). C: steric repulsion between chloroindane in 15b and the side chain
of Lys89 (stereoview). Dark blue stick: the side chain of Lys89, whose conformation originated from X-ray analysis of Thr160
phosphorylated Cdk2-cyclin A-ATPγS complex (PDB ID: 1J ST). The location of Lys89 was determined by superposition of the
Cdk4 homology model and 1J ST. Cyan cloud: van der Waals surface representation of chloroindane. Dark blue cloud: van der
Waals surface representation of Lys89 in Cdk2.
mined with a micromass Q-TOF2 (ESI positive). Optical
rotations were measured with a J asco DIP-370 polarimeter.
Thin-layer chromatography (TLC) analysis was performed on
Merck Kieselgel F254 precoated plates. Silica gel column
chromatography was carried out on Wako gel C-300 (Wako
Pure Chemicals Industries, Ltd., J apan).
The mixture was stirred for 10 h at 50 °C under an atmo-
spheric pressure of hydrogen. The catalyst was removed by
filtration and concentrated in vacuo. The residue was purified
by silica gel column chromatography (3% MeOH in CHCl3) and
recrystalized from EtOAc to yield 2a (58 mg, 53%, two steps)
as a white solid. mp 248-249 °C. 1H NMR (300 MHz, DMSO-
d6): δ 1.02-1.09 (1H, m), 2.26-2.37 (2H, m), 2.59-2.64 (1H,
N-[(9bR)-5-Oxo-2,3,5,9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]-
isoin d ol-9-yl]-N′-(1H-p yr a zol-3-yl)u r ea (2a ). To a stirred
solution of 1-benzyl-3-aminopyrazole24 (150 mg, 0.74 mmol) in
CHCl3 (10 mL) were added 4-(dimethylamino)pyridine (139
mg, 1.63 mmol) and 4-nitrophenyl chloroformate (143 mg, 0.74
mmol). After the solution was stirred for 30 min at 70 °C,
(9bR)-9-amino-1,2,3,9b-tetrahydro-5H-pyrrolo[2,1-a]isoindol-
5-one6 (70 mg, 0.37 mmol) was added, and the reaction mixture
was stirred for 10 h at the same temperature. The solvent was
removed in vacuo, and the residue was purified by silica gel
column chromatography (3% MeOH in CHCl3) to give crude
N-(1-benzyl-1H-pyrazol-3-yl)-N′-[(9bR)-5-oxo-2,3,5,9b-tetrahy-
dro-1H-pyrrolo[2,1-a]isoindol-9-yl]urea (136 mg). The solu-
tion of crude N-(1-benzyl-1H-pyrazol-3-yl)-N′-[(9bR)-5-oxo-
2,3,5,9b-tetrahydro-1H-pyrrolo[2,1-a]isoindol-9-yl]urea (136
mg) in MeOH-tetrahydrofuran (1:1) was mixed with pal-
ladium hydroxide (on carbon, 100 mg) and 2 N HCl (100 µL).
m), 3.26-3.33 (1H, m), 3.52-3.56 (1H, m), 4.70 (1H, dd, J 1
)
10.4 Hz, J 2 ) 5.7 Hz), 6.14 (1H, brs), 7.27 (1H, d, J ) 8.0 Hz),
7.43 (1H, t, J ) 8.0 Hz), 7.64 (1H, s), 8.25 (1H, d, J ) 8.0 Hz),
8.40-9.60 (1H, br), 9.47 (1H, s), 12.39 (1H, brs). MS (ESI+):
m/z 298 (M+H)+. HR-MS calcd for C15H16N5O2 (M+H)+:
20
298.1304, found 298.1300. [R]D 115.8° (c 1.0, CHCl3). Anal.
(C15H15N5O2‚0.1EtOAc) C, H, N.
N-[(9bS)-5-Oxo-2,3,5,9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]-
isoin d ol-9-yl]-N′-(1H-p yr a zol-3-yl)u r ea (2b). To a stirred
solution of 1-benzyl-3-aminopyrazole24 (150 mg, 0.74 mmol)
in CHCl3 (10 mL) were added 4-(dimethylamino)pyridine
(139 mg, 1.63 mmol) and 4-nitrophenyl chloroformate (143 mg,
0.74 mmol). After the solution was stirred for 30 min at 70
°C, (9bS)-9-amino-1,2,3,9b-tetrahydro-5H-pyrrolo[2,1-a]isoin-
dol-5-one6 (70 mg, 0.37 mmol) was added, and the reaction
mixture was stirred for 10 h at the same temperature. The

4636 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26
Honma et al.
F igu r e 6. Cdk4 selectivity of key compounds 1 and 15b over Cdk1/2 and other representative kinases, and sequence alignment
(Asp99, Thr102, and Gln98).
solvent was removed in vacuo, and the residue was purified
by silica gel column chromatography (3% MeOH in CHCl3) to
give crude N-(1-benzyl-1H-pyrazol-3-yl)-N′-[(9bS)-5-oxo-2,3,5,-
9b-tetrahydro-1H-pyrrolo[2,1-a]isoindol-9-yl]urea (149 mg).
The solution of crude N-(1-benzyl-1H-pyrazol-3-yl)-N′-[(9bS)-
5-oxo-2,3,5,9b-tetrahydro-1H-pyrrolo[2,1-a]isoindol-9-yl]urea
(149 mg) in MeOH-tetrahydrofuran (1:1) was mixed with
palladium hydroxide (on carbon, 100 mg) and 2 N HCl (100
µL). The mixture was stirred for 10 h at 50 °C under an
atmospheric pressure of hydrogen. The catalyst was removed
by filtration and concentrated in vacuo. The residue was
purified by silica gel column chromatography (3% MeOH in
CHCl3) and recrystalized from EtOAc to yield 2b (63 mg, 57%,
two steps) as a white solid. mp 248-249 °C. 1H NMR (300
MHz, DMSO-d6): δ 1.02-1.09 (1H, m), 2.26-2.37 (2H, m),
2.59-2.64 (1H, m), 3.26-3.33 (1H, m), 3.52-3.56 (1H, m), 4.70
(1H, dd, J 1 ) 10.4 Hz, J 2 ) 5.7 Hz), 6.14 (1H, brs), 7.27 (1H,
d, J ) 8.0 Hz), 7.43 (1H, t, J ) 8.0 Hz), 7.64 (1H, s), 8.25 (1H,
d, J ) 8.0 Hz), 8.40-9.60 (1H, br), 9.47 (1H, s), 12.39 (1H,
brs). MS (ESI+): m/z 298 (M+H)+. HR-MS calcd for C15H16N5O2
(M+H)+: 298.1304, found 298.1304. [R]D20-115.4° (c 1.0,
CHCl3). Anal. (C15H15N5O2‚0.1EtOAc) C, H, N.
5-[(Ben zyloxy)m eth yl]-1H-p yr a zol-3-a m in e (4). To a
stirred solution of acetonitrile (10.2 mL, 195 mmol) in tet-
rahydrofuran (500 mL) was added butyllithium (130 mL, 1.5
M solution of hexane, 196 mmol) at -78 °C. After the solution
was stirred for 30 min at the same temperature, ethyl benzyl-
oxyacetate (3)25 (34.0 g, 178 mmol) was added. The reaction
mixture was stirred at room temperature for 2 h and then
poured into water. The solution was acidified with 1 N HCl
and extracted with EtOAc. The organic layer was concentrated
in vacuo. A solution of the residue (17.3 g) in ethanol (200 mL)

Development of Selective Cdk4 Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26 4637
was mixed with hydrazine monohydrate (26 mL, 536 mmol)
and refluxed for 10 h. The reaction mixture was evaporated
in vacuo, and the residue was diluted with CHCl3 and washed
with brine. The organic layer was concentrated in vacuo, and
the residue was purified by silica gel column chromatography
(2% MeOH in CHCl3) to give 4 (14.5 g, 40%, two steps). 1H
NMR (200 MHz, CDCl3): δ 3.48 (2H, s), 4.47 (2H, s), 4.53 (2H,
s), 5.58 (1H, s), 7.20-7.44 (6H, m). MS (ESI+): m/z 204
(M+H)+.
MeOH in CHCl3) to yield 8 (0.56 g, 59%) as a white solid. mp
230 °C, dec 1H NMR (300 MHz, DMSO-d6): δ 0.98-1.09 (1H,
m), 2.25-2.36 (2H, m), 2.48-3.36 (2H, m), 3.49-3.53 (1H, m),
4.74 (1H, dd, J 1 ) 10.4 Hz, J 2 ) 5.7 Hz), 7.14 (1H, s), 7.32
(1H, d, J ) 8.0 Hz), 7.46 (1H, t, J ) 8.0 Hz), 8.14 (1H, d, J )
8.0 Hz), 8.42 (1H, brs), 9.68 (1H, br), 9.79 (1H, brs), 10.26 (1H,
s). MS (ESI+): m/z 326 (M+H)+. HR-MS calcd for C16H16N5O3
(M+H)+: 326.1253, found 326.1262.
Gen er a l P r oced u r e for P r ep a r a tion of Alk yla m in om -
eth yl Libr a r y (9). To a stirred solution of 8 (30 mg, 0.092
mmol) and MS3Å (30 mg) in MeOH (3 mL) was added 3.0 equiv
of amine at room temperature. After the solution was stirred
for 12 h at the same temperature, NaBH4 (100 mg, 2.64 mmol)
was added. The resultant mixture was stirred for 1 h, poured
into 1 N HCl, neutralized with saturated aqueous NaHCO3,
and extracted with CHCl3. The organic layer was washed with
brine, dried over MgSO4, and concentrated in vacuo. The
residue was purified by preparative silica gel chromatography
(20% MeOH in CHCl3) to yield 9. Yields ranged from 20 to
50%.
ter t-Bu tyl 3-Am in o-5-[(ben zyloxy)m eth yl]-1H-pyr azole-
1-ca r boxyla te (5). To a stirred solution of 4 (14.5 g, 71.3
mmol) in tetrahydrofuran-dimethylformamide (3:2, 250 mL)
was added sodium hydride (3.1 g, 60 wt % in mineral oil, 78.4
mmol) at 0 °C. After the solution was stirred for 30 min, di-
tert-butyl dicarbonate (17.1 g, 78.4 mmol) was added. The
resultant mixture was stirred for 2 h at room temperature,
poured into saturated aqueous NH4Cl, and extracted with
CHCl3. The organic layer was washed with brine, dried over
MgSO4, and concentrated in vacuo. The residue was purified
by silica gel column chromatography (2% MeOH in CHCl3) to
1
The following compounds (9a , 9b, 9c, 9d , 9e, 9f, 9g, 9h ,
15a , and 15b) were prepared from the appropriate amines and
8 by the same procedure.
yield 5 (11.5 g, 53%). H NMR (300 MHz, CDCl3): δ 1.68 (9H,
s), 3.49 (1H, d, J ) 5.6 Hz), 4.52 (2H, s), 4.58 (2H, s), 5.34
(2H, br), 5.54 (1H, s), 7.28-7.40 (5H, m). MS (ESI+): m/z 304
(M+H)+.
N-{5-[(Met h yla m in o)m et h yl]-1H -p yr a zol-3-yl}-N′-(5-
oxo-2,3,5,9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]isoin d ol-9-yl)-
N-{5-[(Ben zyloxy)m et h yl]-1H-p yr a zol-3-yl}-N′-(5-oxo-
2,3,5,9b-t et r a h yd r o-1H -p yr r olo[2,1-a ]isoin d ol-9-yl)u r ea
(6). To a stirred solution of 5 (5.0 g, 16 mmol) in CHCl3 (50
mL) were added 4-(dimethylamino)pyridine (2.2 g, 18 mmol)
and 4-nitrophenyl chloroformate (3.65 g, 17 mmol). After the
solution was stirred for 30 min at 70 °C, 9-amino-1,2,3,9b-
tetrahydro-5H-pyrrolo[2,1-a]isoindol-5-one6 (3.4 g, 17 mmol)
was added. The reaction mixture was stirred for 2 h at the
same temperature. The solvent was removed in vacuo, and the
residue was purified by silica gel column chromatography (2%
MeOH in CHCl3) to give crude tert-butyl 5-[(benzyloxy)methyl]-
3-({[(5-oxo-2,3,5,9b-tetrahydro-1H-pyrrolo[2,1-a]isoindol-9-yl)-
amino]carbonyl}amino)-1H-pyrazole-1-carboxylate (2.5 g). A
solution of crude tert-butyl 5-[(benzyloxy)methyl]-3-({[(5-oxo-
2,3,5,9b-tetrahydro-1H-pyrrolo[2,1-a ]isoindol-9-yl)amino]-
carbonyl}amino)-1H-pyrazole-1-carboxylate (2.5 g) in MeOH
was mixed with 4 N HCl (5 mL, dioxane solution). After the
solution was stirred for 30 min at room temperature, the
reaction mixture was diluted with CHCl3, and the organic layer
was washed with brine, dried over MgSO4, and concentrated
in vacuo. The residue was purified by silica gel column
chromatography (5% MeOH in CHCl3) to yield 6 (1.3 g, 19%,
1
u r ea (9a ). H NMR (300 MHz, DMSO-d6): δ 0.95-1.10 (1H,
m), 2.49 (3H, s), 2.20-2.80 (3H, m), 3.20-3.60 (2H, m), 4.12
(2H, m), 4.79 (1H, dd, J 1 ) 10.2 Hz, J 2 ) 5.3 Hz), 6.41 (1H, s),
7.29 (1H, d, J ) 7.9 Hz), 7.44 (1H, t, J ) 7.9 Hz), 8.21 (1H, d,
J ) 7.9 Hz), 9.12 (2H, br), 9.90 (1H, brs). MS (ESI+): m/z 341
(M+H)+. HR-MS calcd for C17H21N6O2 (M+H)+: 341.1726,
found 341.1726.
N-(5-Oxo-2,3,5,9b-tetr ah ydr o-1H-pyr r olo[2,1-a ]isoin dol-
9-yl)-N ′-{5-[(p r o p yla m in o)m e t h yl]-1H -p y r a zo l-3-y l}-
1
u r ea (9b). H NMR (300 MHz, DMSO-d6): δ 0.91 (1H, t, J )
7.5 Hz), 0.90-1.22 (1H, m), 1.60-1.75 (1H, m), 2.22-2.40 (2H,
m), 2.65-2.77 (1H, m), 2.80-2.82 (2H, m), 3.25-3.35 (1H,m),
3.45-3.60 (1H, m), 4.13 (1H, m), 4.80 (1H, dd, J 1 ) 10.4 Hz,
J 2 ) 5.9 Hz), 6.43 (1H, s), 7.28 (1H, d, J ) 7.3 Hz), 7.44 (1H,
t, J ) 7.3 Hz), 8.21 (1H, d, J ) 7.3 Hz), 9.23 (2H, br), 9.95
(1H, s). MS (ESI+): m/z 369 (M+H)+. HR-MS calcd for
C
19H25N6O2 (M+H)+: 369.2039, found 369.2036.
N-{5-[(Isop r op yla m in o)m eth yl]-1H-p yr a zol-3-yl}-N′-(5-
oxo-2,3,5,9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]isoin d ol-9-yl)-
1
u r ea (9c). H NMR (300 MHz, DMSO-d6): δ 0.91 (1H, t, J )
7.5 Hz), 0.98-1.15 (1H, m), 1.27 (6H, d, J ) 6.4 Hz), 2.25-
2.42 (2H, m), 2.40-2.60 (1H, m), 2.66-2.78 (1H, m), 3.25-
3.42 (2H, m), 3.45-3.62 (1H, m), 4.13 (2H, m), 4.80 (1H, dd,
J 1 ) 10.4 Hz, J 2 ) 5.9 Hz), 6.44 (1H, s), 7.43 (1H, d, J ) 7.7
Hz), 8.20 (1H, d, J ) 7.7 Hz), 9.15 (2H, br), 9.91 (1H, s). MS
(ESI+): m/z 369 (M+H)+. HR-MS calcd for C19H25N6O2
(M+H)+: 369.2039, found 369.2039.
1
two steps) as a white solid. H NMR (300 MHz, DMSO-d6): δ
1.01-1.10 (1H, m), 2.28-2.33 (2H, m), 2.59-2.65 (1H, m),
3.25-3.31 (1H, m), 3.51 (1H, d, J ) 8.2 Hz), 4.50 (2H, s), 4.70
(1H, dd, J 1 ) 10.5 Hz, J 2 ) 5.9 Hz), 6.15 (1H, brs), 7.26-7.46
(8H, m), 8.25 (1H, d, J ) 8.2 Hz), 9.46 (1H, s), 12.50 (1H, brs).
MS (ESI+): m/z 418 (M+H)+.
N-{5-[(ter t-Bu tyla m in o)m eth yl]-1H-p yr a zol-3-yl}-N′-(5-
oxo-2,3,5,9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]isoin d ol-9-yl)-
N-[5-(Hydr oxym eth yl)-1H-pyr azol-3-yl]-N′-(5-oxo-2,3,5,-
9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]isoin d ol-9-yl)u r ea (7). To
a stirred solution of 6 (1.3 g, 3.1 mmol) in MeOH-tetrahy-
drofuran (1:1, 100 mL) were added palladium hydroxide (on
carbon, 500 mg) and 2 N HCl (1.0 mL), and the mixture was
stirred for 10 h at 50 °C under an atmospheric pressure of
hydrogen. The catalyst was removed by filtration and concen-
trated in vacuo. The residue was precipitated from EtOAc-
1
u r ea (9d ). H NMR (300 MHz, DMSO-d6): δ 0.91 (1H, t, J )
7.5 Hz), 0.95-1.12 (1H, m), 1.35 (9H, s), 2.22-2.38 (2H, m),
2.62-2.75 (1H, m), 3.32-3.37 (1H, m), 3.42-3.60 (1H, m), 4.10
(2H, m), 4.79 (1H, dd, J 1 ) 10.4 Hz, J 2 ) 5.9 Hz), 6.47 (1H, s),
7.29 (1H, d, J ) 7.3 Hz), 7.45 (1H, t, J ) 7.3 Hz), 8.22 (1H, d,
J ) 7.3 Hz), 9.09 (2H, br), 9.91 (1H, s). MS (ESI+): m/z 383
(M+H)+. HR-MS calcd for C20H27N6O2 (M+H)+: 383.2195,
found 383.2200.
1
hexane to yield 7 (0.94 g, 93%) as a white solid. H NMR (300
MHz, DMSO-d6): δ 1.01-1.08 (1H, m), 2.30-2.33 (2H, m),
N-{5-[(Cyclop en tyla m in o)m eth yl]-1H-p yr a zol-3-yl}-N′-
(5-oxo-2,3,5,9b -t et r a h yd r o-1H -p yr r olo[2,1-a ]isoin d ol-9-
yl)u r ea (9e). 1H NMR (300 MHz, DMSO-d6): δ 0.90-1.20 (1H,
m), 1.20-2.00 (8H, m), 2.20-2.70 (4H, m), 3.00-3.40 (1H, m),
3.40-3.60 (1H, m), 3.74 (2H, m), 4.69 (1H, m), 7.25 (1H, d, J
) 7.9 Hz), 7.41 (1H, t, J ) 7.9 Hz), 8.21 (1H, d, J ) 7.9 Hz),
9.44 (1H, br), 12.20 (1H, br). MS (ESI+): m/z 395 (M+H)+.
2.63-2.66 (1H, m), 3.16 (1H, d, J ) 5.2 Hz), 3.47-3.53 (1H,
m), 4.64 (2H, d, J ) 5.6 Hz), 4.70 (1H, dd, J 1 ) 10.3 Hz, J 2
)
5.7 Hz), 5.30 (1H, d, J ) 5.4 Hz), 6.02 (1H, brs), 7.26 (1H, d,
J ) 6.6 Hz), 7.43 (1H, t, J ) 6.6 Hz), 8.27 (1H, d, J ) 6.6 Hz),
9.44 (1H, s), 12.30 (1H, brs). MS (ESI+): m/z 328 (M+H)+.
N-(5-For m yl-1H-pyr azol-3-yl)-N′-(5-oxo-2,3,5,9b-tetr ah y-
d r o-1H-p yr r olo[2,1-a ]isoin d ol-9-yl)u r ea (8). To a stirred
solution of 7 (0.94 g, 2.9 mmol) in CHCl3-dimethylformamide
(1:2, 15 mL) was added MnO2 (1.5 g, 17.3 mmol). After the
solution was stirred for 1 h at 50 °C, the reaction mixture was
filtered. The filtrate was concentrated in vacuo, and the
residue was purified by silica gel column chromatography (10%
HR-MS calcd for
395.2205.
C
21H27N6O2 (M+H)+: 395.2195, found
N-{5-[(Cycloh exyla m in o)m eth yl]-1H-p yr a zol-3-yl}-N′-
(5-oxo-2,3,5,9b -t et r a h yd r o-1H -p yr r olo[2,1-a ]isoin d ol-9-
yl)u r ea (9f). 1H NMR (300 MHz, DMSO-d6): δ 0.80-1.45 (6H,
m), 1.58-1.69 (1H, m), 1.70-1.85 (2H, m), 2.05-2.15 (2H, m),

4638 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26
Honma et al.
2.20-2.40 (2H, m), 2.65-2.75 (1H, m), 2.92-3.07 (1H, m),
3.24-3.38 (1H, m), 3.45-3.60 (1H, m), 4.17 (1H, m), 4.79 (1H,
dd, J 1 ) 10.4 Hz, J 2 ) 5.9 Hz), 6.45 (1H, s), 7.29 (1H, d, J )
7.5 Hz), 7.45 (1H, t, J ) 7.5 Hz), 8.26 (1H, d, J ) 7.5 Hz), 9.14
(2H, br), 9.89 (1H, s). MS (ESI+): m/z 409 (M+H)+. HR-MS
calcd for C22H29N6O2 (M+H)+: 409.2352, found 409.2344.
was added. The reaction mixture was stirred for 12 h at 60
°C, poured into 1 N NaOH, and extracted with CHCl3. The
organic layer was washed with brine, dried over MgSO4, and
concentrated in vacuo. The residue was purified by silica gel
chromatography (1.5% MeOH in CHCl3) to give tert-butyl
5-[(2S)-1-benzylpyrrolidin-2-yl]-3-({[(5-oxo-2,3,5,9b-tetrahydro-
1H-pyrrolo[2,1-a]isoindol-9-yl)amino]carbonyl}amino)-1H-pyra-
N-(5-Oxo-2,3,5,9b-tetr ah ydr o-1H-pyr r olo[2,1-a ]isoin dol-
9-yl)-N ′-[5-(p yr r o lid in -1-y lm e t h y l)-1H -p y r a zo l-3-y l]-
1
zole-1-carboxylate (1.55 g, 42%). H NMR (300 MHz, DMSO-
1
d6): δ 1.01-1.23 (1H, m), 1.64-1.80 (3H, m), 1.65 and 1.69
(each 4.5H, s), 2.00-2.58 (5H, m), 2.59-2.90 (1H, m), 3.17 (1H,
d, J ) 13.0 Hz), 3.43-3.62 (3H, m), 3.78 (1H, d, J ) 13.0 Hz),
4.64-4.70 (1H, m), 6.43 (1H, brs), 7.20-7.36 (6H, m), 7.43 (1H,
t, J ) 8.0 Hz), 7.55 (1H, d, J ) 8.0 Hz), 8.35 (1H, d, J ) 8.0
Hz), 8.70-9.04 (1H, m), 9.96-10.51 (1H, m). MS (ESI+): m/z
557 (M+H)+.
u r ea (9g). H NMR (300 MHz, DMSO-d6): δ 0.95-1.10 (1H,
m), 1.75-2.10 (4H, m), 2.15-2.40 (2H, m), 2.66-2.78 (1H, m),
3.00-3.20 (2H, m), 3.20-3.65 (4H, m), 4.32 (2H, m), 4.79 (1H,
dd, J 1 ) 10.3 Hz, J 2 ) 5.3 Hz), 6.43 (1H, s), 7.28 (1H, d, J )
7.9 Hz), 7.43 (1H, t, J ) 7.9 Hz), 8.19 (1H, d, J ) 7.9 Hz), 9.10
(1H, br), 9.90 (1H, br), 10.63 (1H, br). MS (ESI+): m/z 381
(M+H)+. HR-MS calcd for C20H25N6O2 (M+H)+: 381.2039,
found 381.2044.
N-{5-[(2S)-1-Ben zylp yr r olid in -2-yl]-1H -p yr a zol-3-yl}-
N′-(5-oxo-2,3,5,9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]isoin d ol-
9-yl)u r ea (13). tert-Butyl 5-[(2S)-1-benzylpyrrolidin-2-yl]-3-
({[(5-oxo-2,3,5,9b-tetrahydro-1H-pyrrolo[2,1-a ]isoindol-9-
yl)amino]carbonyl}amino)-1H-pyrazole-1-carboxylate (1.55 g,
2.8 mmol) was diluted with 4 N HCl (MeOH solution, 20 mL),
and the resultant solution was stirred for 10 h at room
temperature. The mixture was poured into saturated aqueous
NaHCO3, extracted with CHCl3, washed with brine, dried over
MgSO4, and concentrated in vacuo. The residue was purified
by silica gel column chromatography (1.5% MeOH in CHCl3)
N-(5-Oxo-2,3,5,9b-tetr ah ydr o-1H-pyr r olo[2,1-a ]isoin dol-
9-yl)-N′-[5-(p ip er id in -1-ylm et h yl)-1H -p yr a zol-3-yl]u r ea
(9h ). 1H NMR (300 MHz, DMSO-d6): δ 0.80-2.00 (7H, m),
2.15-2.95 (3h, m), 3.20-3.70 (6H, m), 4.25 (2H, m), 4.74 (1H,
dd, J 1 ) 10.3 Hz, J 2 ) 5.3 Hz), 6.43 (1H, s), 7.27 (1H, d, J )
7.9 Hz), 7.42 (1H, t, J ) 7.9 Hz), 8.17 (1H, d, J ) 7.9 Hz), 8.90
(1H, br), 9.75 (1H, br). MS (ESI+): m/z 395 (M+H)+. HR-MS
calcd for C21H27N6O2 (M+H)+: 395.2195, found 395.2196.
5-[(2S)-1-Ben zylp yr r olid in -2-yl]-1H -p yr a zol-3-a m in e
(11). To a stirred solution of acetonitrile (0.784 mL, 15.0 mmol)
in tetrahydrofuran (25 mL) was added butyllithium (9.6 mL,
1.56 M solution of hexane, 15.0 mmol) at -78 °C. After the
solution was stirred for 30 min at the same temperature,
N-benzyl-L-proline ethyl ester (10) (3.18 g, 13.6 mmol) was
added. The reaction mixture was stirred at room temperature
for 2 h and then poured into water. The solution was acidified
by 1 N HCl, neutralized by saturated aqueous NaHCO3, and
extracted with EtOAc. The organic layer was concentrated in
vacuo. A solution of the residue (2.73 g) in ethanol (30 mL)
was mixed with hydrazine monohydrate (3.0 mL, 61.8 mmol),
and the resultant mixture was refluxed for 24 h. The reaction
mixture was evaporated in vacuo, and the residue was diluted
with CHCl3 and washed with brine. The organic layer was
concentrated in vacuo, and the residue was purified by silica
gel column chromatography (5% MeOH in CHCl3) to give 11
1
to yield 13 (623 mg, 21%). H NMR (300 MHz, DMSO-d6): δ
0.98-1.04 (1H, m), 1.64-1.80 (3H, m), 2.04-2.40 (4H, m),
2.59-2.90 (2H, m), 3.16 (1H, d, J ) 13.2 Hz), 3.42-3.60 (3H,
m), 3.76 (1H, d, J ) 13.2 Hz), 4.62-4.68 (1H, m), 6.09 (1H,
brs), 7.20-7.36 (6H, m), 7.42 (1H, t, J ) 7.9 Hz), 8.26 (1H, d,
J ) 7.9 Hz), 9.43 (1H, s), 12.40 (1H, s). MS (ESI+): m/z 457
(M+H)+. HR-MS calcd for C26H29N6O2 (M+H)+: 457.2352,
20
found 457.2352. [R]D -46.8° (c 1.0, dimethylformamide).
N-[(9bR)-5-Oxo-2,3,5,9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]i-
soin d ol-9-yl]-N′-{5-[(2S)-p yr r olid in -2-yl]-1H -p yr a zol-3-
yl}u r ea (14a ) a n d N-[(9bS)-5-Oxo-2,3,5,9b-tetr a h yd r o-1H-
p yr r olo[2,1-a ]isoin d ol-9-yl]-N′-{5-[(2S)-p yr r olid in -2-yl]-
1H-p yr a zol-3-yl}u r ea (14b). To a stirred solution of 13 (620
mg, 1.4 mmol) in MeOH (5.0 mL) were added 4 N HCl (MeOH
solution, 0.5 mL) and 10% Pd/C (250 mg). The mixture was
stirred at room temperature for 3 days under an atmospheric
pressure of hydrogen. The catalyst was removed by filtration,
and the filtrate was concentrated in vacuo. The residue was
recrystalized from EtOAc-MeOH to give a diastereomeric
mixture 14 (264 mg, 51%). A sample of 14 (169 mg) was
separated by chiral HPLC (CHIRALPAK AD, 20φ × 250 mm,
ethyl alcohol-ethylamine (1000:1), flow rate 10 mL/min) to
give 14a (retention time 10 min, 45 mg, 99% de) and 14b
(retention time 9 min, 28 mg, 99% de) as white solids. 14a :
1
(2.65 g, 81%) as a colorless foam. H NMR (300 MHz, CDCl3):
δ 1.98-2.08 (3H, m), 2.36-2.50 (2H, m), 3.24-3.30 (1H, m),
3.46 (1H, d, J ) 12.9 Hz), 3.80 (1H, t, J ) 7.3 Hz), 3.70-4.00
(2H, br), 4.15 (1H, d, J ) 12.9 Hz), 5.83 (1H, s), 7.48-7.57
(5H, m). MS (ESI+): m/z 243 (M+H)+. [R]D -70.2° (c 1.0,
20
dimethylformamide).
ter t-Bu tyl 3-Am in o-5-[(2S)-1-ben zylp yr r olid in -2-yl]-1H-
p yr a zole-1-ca r boxyla te (12). To a stirred solution of 11 (2.65
g, 10.9 mmol) in dimethylformamide (40 mL) was added NaH
(60 wt % in mineral oil, 0.52 g, 13.0 mmol). After the solution
was stirred for 30 min at room temperature, di-tert-butyl
dicarbonate (3.0 mL, 13.1 mmol) was added. The reaction
mixture was stirred for 30 min, poured into saturated aqueous
NH4Cl, and extracted with EtOAc. The organic layer was
washed with water and brine, dried over MgSO4, and concen-
trated in vacuo. The residue was purified by silica gel column
chromatography (EtOAc 100%) to give 12 (2.26 g, 61%) as a
colorless foam. 1H NMR (300 MHz, CDCl3): δ 1.65 (9H, s),
1.74-1.86 (3H, m), 2.12-2.20 (2H, m), 2.86-3.02 (1H, m), 3.13
(1H, d, J ) 13.1 Hz), 3.47 (1H, t, J ) 7.9 Hz), 3.93 (1H, d, J )
12.9 Hz), 5.24 (2H, brs), 5.57 (1H, s), 7.19-7.31 (5H, m). MS
1
mp 223-225 °C. H NMR (300 MHz, DMSO-d6): δ 0.98-1.18
(1H, m), 1.59-1.80 (3H, m), 2.00-2.10 (1H, m), 2.22-2.40 (2H,
m), 2.60-2.77 (1H, m), 2.80-2.98 (2H, m), 3.42-3.48 (1H, m),
4.09 (1H, t, J ) 7.2 Hz), 4.64-4.79 (1H, m), 5.86 (1H, s), 7.26
(1H, d, J ) 6.7 Hz), 7.42 (1H, dd, J 1 ) 8.0 Hz, J 2 ) 6.7 Hz),
8.27 (1H, d, J ) 8.0 Hz), 9.41 (1H, s), 12.19 (1H, s). MS
(ESI+): m/z 367 (M+H)+. HR-MS calcd for C19H23N6O2
20
(M+H)+: 367.1882, found 367.1895. [R]D 53.2° (c 0.5, dim-
1
ethylformamide). 14b: mp 229-231 °C. H NMR (300 MHz,
DMSO-d6): δ 0.98-1.18 (1H, m), 1.59-1.80 (3H, m), 2.00-
2.10 (1H, m), 2.22-2.40 (2H, m), 2.60-2.77 (1H, m), 2.80-
2.98 (2H, m), 3.42-3.48 (1H, m), 4.09 (1H, t, J ) 7.2 Hz), 4.64-
4.79 (1H, m), 5.86 (1H, s), 7.26 (1H, d, J ) 6.7 Hz), 7.42 (1H,
dd, J 1 ) 8.0 Hz, J 2 ) 6.7 Hz), 8.27 (1H, d, J ) 8.0 Hz), 9.41
(1H, s), 12.19 (1H, s). MS (ESI+): m/z 367 (M+H)+. HR-MS
(ESI+): m/z 343 (M+H)+. [R]D -21.8° (c 1.0, dimethylfor-
20
mamide).
ter t-Bu tyl 5-[(2S)-1-Ben zylp yr r olid in -2-yl]-3-({[(5-oxo-
2,3,5,9b-tetr ah ydr o-1H-pyr r olo[2,1-a ]isoin dol-9-yl)am in o]-
ca r bon yl}a m in o)-1H-p yr a zole-1-ca r boxyla te. To a stirred
solution of 12 (2.26 g, 6.6 mmol) in tetrahydrofuran (30 mL)
was added NaH (60 wt % in mineral oil, 344 mg, 8.6 mmol).
After the solution was stirred for 30 min at room temperature,
4-nitrophenyl chloroformate (1.73 g, 8.6 mmol) was added and
stirred for an additional 30 min; subsequently, 4-(dimethy-
lamino)pyridine (960 mg, 7.9 mmol) and 9-amino-1,2,3,9b-
tetrahydro-5H-pyrrolo[2,1-a]isoindol-5-one25 (1.62 g, 8.6 mmol)
calcd for C19H23N6O2 (M+H)+: 367.1882, found 367.1892. [R]D
-78.4° (c 0.5, dimethylformamide).
20
(+)-N-[(2S)-5-Ch lor o-2,3-d ih yd r o-1H -in d en -2-yl]-4-n i-
tr oben za m id e a n d (-)-N-[(2R)-5-Ch lor o-2,3-d ih yd r o-1H-
in d en -2-yl]-4-n itr oben za m id e. To a solution of 2-aminoin-
dan (14.8 g, 111 mmol) in acetic acid (200 mL) was added Cl2
gas by bubbling for 5 min at room temperature. After the
solution was stirred for 15 min, Cl2 was removed by the
bubbling of N2 gas. The solution was concentrated in vacuo,

Development of Selective Cdk4 Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26 4639
and the residue was washed with hexane-EtOAc (2:1) to yield
crude 5-chloro-2,3-dihydro-1H-inden-2-ylamine (23 g). To a
solution of the crude 5-chloro-2,3-dihydro-1H-inden-2-ylamine
(23 g) in CHCl3 was added triethylamine (31 mL, 220 mmol)
at room temperature. Subsequently, p-nitrobenzoyl chloride
(22 g, 120 mmol) was added. After the solution was stirred
for 2 h, the reaction mixture was poured into saturated
aqueous NaHCO3 solution. The aqueous layer was extracted
with CHCl3. The combined organic layer was washed with
brine, dried over MgSO4, and concentrated in vacuo. The
residue was purified by HPLC (Senshu Pak PEGASIL Silica
60-5, 30φ × 250 mm, hexane-EtOAc (1:1), flow rate 15 mL/
min) to give the racemate N-(5-chloro-2,3-dihydro-1H-inden-
2-yl)-4-nitrobenzamide (retention time 13 min, 7.8 g, 22%).
N-(5-chloro-2,3-dihydro-1H-inden-2-yl)-4-nitrobenzamide (7.8
g) was separated by chiral HPLC (CHIRALCEL OD, 20φ ×
250 mm, hexane-isopropyl alcohol (2:3), flow rate 10 mL/min)
to yield (+)-N-[(2S)-5-chloro-2,3-dihydro-1H-inden-2-yl]-4-ni-
trobenzamide (retention time 17 min, 3.63 g, 99% ee) and (-)-
N-[(2R)-5-chloro-2,3-dihydro-1H-inden-2-yl]-4-nitrobenza-
mide (retention time 14 min, 3.76 g, 99% ee) as white solids.
2.80 (3H, m), 2.98-3.18 (2H, m), 3.22-3.60 (3H, m), 3.71 (2H,
s), 4.63-4.78 (1H, m), 6.08 (1H, s), 7.08-7.22 (3H, m), 7.24
(1H, d, J ) 6.5 Hz), 7.44 (1H, dd, J 1 ) 7.9 Hz, J 2 ) 6.5 Hz),
8.27 (1H, d, J ) 7.9 Hz), 9.43 (1H, s), 12.22 (1H, s). MS
(ESI+): m/z 477 (M+H)+. HR-MS calcd for C25H26N6O2Cl
20
(M+H)+: 477.1806, found 477.1800. [R]D 72.2° (c 1.0, dim-
ethylformamide). Anal. (C25H25N6O2Cl‚1H2O) C, H, N. 15b: mp
215-217 °C. 1H NMR (300 MHz, DMSO-d6): δ 0.98-1.14 (1H,
m), 2.23-2.40 (1H, m), 2.59-2.80 (3H, m), 2.98-3.18 (2H, m),
3.22-3.60 (3H, m), 3.71 (2H, s), 4.63-4.78 (1H, m), 6.08 (1H,
s), 7.08-7.22 (3H, m), 7.24 (1H, d, J ) 6.5 Hz), 7.44 (1H, dd,
J 1 ) 7.9 Hz, J 2 ) 6.5 Hz), 8.27 (1H, d, J ) 7.9 Hz), 9.43 (1H,
s), 12.22 (1H, s). MS (ESI+): m/z 477 (M+H)+. HR-MS calcd
for C25H26N6O2Cl (M+H)+: 477.1806, found 477.1805. [R]D
20
-62.4° (c 1.0, dimethylformamide). Anal. (C25H25N6O2Cl‚1H2O)
C, H, N.
Molecu la r Mod elin g. Computer calculations were per-
formed on a Silicon Graphics Octane. Molecular mechanics
calculations were carried out using the CHARMm force field.
The distance-dependent dielectric constant ꢀ ) 1 was used.
The Docking studies were performed using modeling software
QUANTA (QUANTA Modeling Environment, December 1998,
San Diego: Molecular Simulations Inc., 1998). Initial docking
models between the Cdk2/4 and inhibitors (1, 2a , 9d , and 15b)
were achieved by manual docking based on the X-ray struc-
tures of Cdk2-1 complex and Cdk4 mimic Cdk2-14b com-
plex.9 Subsequently, 200-400 conformations were generated
using random sampling, following minimization for each initial
model. These conformations were clustered using the RMSD
of atomic coordinates (ca. 15-30 clusters). The most preferable
conformations were selected by considering energy, status of
hydrogen-bonds, and van der Waals contacts between hydro-
phobic surfaces. Candidates of the substituent on the pyrazole
ring of 2a were generated by the de novo design programs,
LUDI19 and LeapFrog.20 The docking model of Cdk4-2a was
used as a seed structure for de novo design, and the 5-position
of the pyrazole ring in 2a was selected as a substitution point.
(+)-N-[(2S)-5-Ch lor o-2,3-d ih yd r o-1H -in d en -2-yl]-4-n i-
tr oben za m id e: mp 168-170 °C. 1H NMR (300 MHz, DMSO-
d6): δ 2.88 (1H, t, J ) 5.2 Hz), 2.97 (1H, t, J ) 5.2 Hz), 3.39
(1H, dd, J 1 ) 6.0 Hz, J 2 ) 5.2 Hz), 3.47 (1H, dd, J 1 ) 6.0 Hz,
J 2 ) 5.2 Hz), 4.88-5.05 (1H, m), 6.29-6.42 (1H, m), 7.17-
7.31 (3H, m), 7.89 (1H, d, J ) 8.6 Hz), 8.27 (1H, d, J ) 8.6
Hz). HR-MS calcd for C16H14N2O335Cl (M+H)+: 317.0693,
20
found 317.0696. [R]D 52.4° (c 1.0, CHCl3).
(-)-N-[(2R)-5-Ch lor o-2,3-d ih yd r o-1H -in d en -2-yl]-4-n i-
tr oben za m id e: mp 168-170 °C. 1H NMR (300 MHz, DMSO-
d6): δ 2.88 (1H, t, J ) 5.2 Hz), 2.97 (1H, t, J ) 5.2 Hz), 3.39
(1H, dd, J 1 ) 6.0 Hz, J 2 ) 5.2 Hz), 3.47 (1H, dd, J 1 ) 6.0 Hz,
J 2 ) 5.2 Hz), 4.88-5.05 (1H, m), 6.29-6.42 (1H, m), 7.17-
7.31 (3H, m), 7.89 (1H, d, J ) 8.6 Hz), 8.27 (1H, d, J ) 8.6
Hz). HR-MS calcd for C16H14N2O335Cl (M+H)+: 317.0693,
20
found 317.0698. [R]D -50.2° (c 1.0, CHCl3).
N-[(9bR)-5-Oxo-2,3,5,9b-tetr a h yd r o-1H-p yr r olo[2,1-a ]-
is o in d o l-9-y l]-N ′-[5-({[(2S )-5-c h lo r o -2,3-d ih y d r o -1H -
in den -2-yl]am in o}m eth yl)-1H-pyr azol-3-yl]u r ea (15a) an d
N-[(9b S)-5-Oxo-2,3,5,9b -t e t r a h yd r o-1H -p yr r olo[2,1-a ]-
isoin dol-9-yl]-N′-[5-({[(2S)-5-ch lor o-2,3-dih ydr o-1H-in den -
2-yl]a m in o}m eth yl)-1H-p yr a zol-3-yl]u r ea (15b). To a solu-
tion of (+)-N-[(2S)-5-chloro-2,3-dihydro-1H-inden-2-yl]-4-ni-
trobenzamide (1.52 g, 4.79 mmol) in 12 N HCl (160 mL) was
added acetic acid (30 mL). The resultant mixture was refluxed
for 10 h. After cooling to room temperature, the reaction
mixture was diluted with CHCl3 and extracted with water.
The combined aqueous layer was made alkaline with 12 N
NaOH to pH 11 and extracted with CHCl3. The combined
organic layer was washed by brine, dried over MgSO4, and
concentrated in vacuo to produce crude (2S)-5-chloro-2,3-
dihydro-1H-inden-2-amine (0.29 g, 36%). A solution of 8 (0.50
g, 1.5 mmol) in CHCl3-MeOH (1:1, 10 mL) was mixed with
crude (2S)-5-chloro-2,3-dihydro-1H-inden-2-amine (0.29 g, 1.7
mmol). The resultant mixture was stirred for 10 h at 50 °C.
After the solution was cooled to room temperature, NaBH4 (87
mg, 2.3 mmol) was added, and the mixture was stirred for 1
h. The mixture was poured into saturated aqueous NH4Cl and
extracted with CHCl3. The combined organic layer was washed
with brine, dried over MgSO4, and concentrated in vacuo. The
residue was purified by preparative silica gel chromatography
(Merck Art.5744, 10% MeOH in CHCl3) to give a diastereo-
meric mixture of N-(5-oxo-2,3,5,9b-tetrahydro-1H-pyrrolo[2,1-
a]isoindol-9-yl)-N′-[5-({[(2S)-5-chloro-2,3-dihydro-1H-inden-2-
yl]amino}methyl)-1H-pyrazol-3-yl]urea (0.25 g, 34%) as a white
solid. N-(5-Oxo-2,3,5,9b-tetrahydro-1H-pyrrolo[2,1-a]isoindol-
9-yl)-N′-[5-({[(2S)-5-chloro-2,3-dihydro-1H-inden-2-yl]amino}-
methyl)-1H-pyrazol-3-yl]urea (0.25 g) was separated by chiral
HPLC (CHIRALPAK AD, 20φ × 250 mm, ethyl alcohol 100%,
flow rate 15 mL/min) to yield 15a (retention time 15 min, 109
mg, 100% de) and 15b (retention time 22 min, 113 mg, 100%
Cd k Assa ys (Cyclin D2-Cd k 4, Cyclin D1-Cd k 6, Cyclin
A-Cd k 2, a n d Cyclin B-Cd k 1). In vitro Cdk assays were
performed as previously described26 with some modifications.
A recombinant human cyclin D2-Cdk4, cyclin D1-Cdk6, or
cyclin A-Cdk2 complexes were expressed in Sf9 cells with a
baculovirus expression system and purified using HPLC.
Recombinant human cyclin B-Cdk1 was purchased from
Promega Corporation (USA). The purified Cdks were incubated
with 50 µM ATP (0.5-1.5 µCi of [33P]-ATP (3000 Ci/mmol)),
100 µM or 400 µM G1 peptide (for Cdk4/6) or 10 mg/mL of S1
peptide (for Cdk1/2), R buffer (20 mM Tris-HCl pH 7.4, 10 mM
MgCl2, 4.5 mM 2-mercaptoethanol, 1 mM EGTA), and diluted
compounds at 30 °C for 20 min (for Cdk6) or 45 min (for Cdk1/
2/4). The reactions were terminated by the addition of 350 mM
H3PO4. Peptides were trapped and washed using P81 paper
filter 96-well plates (Millipore) and 75 mM H3PO4. 33P incor-
poration was measured with a Top Count (Packard).
Oth er Ser /Th r Kin a se Assa ys. PKA, PKC, PKBR, CaMK
II, p38R, ERK1, and MEK1 were assayed in MDS pharma
Services.27
Tyr Kin a se Assa ys. Src, Lck, Flt-1, ZAP, EGFR, FGFR1,
and PFGFRâ were assayed in the Merck research laboratory.28
Ack n ow led gm en t. It is our pleasure to acknowl-
edge the contributions of Dr. Toshiyasu Shimomura and
of Mr. Takumitsu Machida (for cell assay), Miss Hiroko
Oki (for Cdk4 inhibitory assay), Mr. Hirokazu Ohsawa
(for mass spectral analysis), Mrs. Chihiro Sato (for
elemental analysis), and Dr. Kenji Kamata (for X-ray
analysis). We thank Mr. Toshiharu Iwama, Dr. Hiroshi
Funabashi, Dr. Hiroshi Hirai, and Dr. Takehiro Fukami
for their useful suggestions. We are also grateful to Miss
J ocelyn Winward for her critical reading of this manu-
script.
1
de) as white solids. 15a : mp 214-215 °C. H NMR (300 MHz,
DMSO-d6): δ 0.98-1.14 (1H, m), 2.23-2.40 (1H, m), 2.59-

4640 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 26
Honma et al.
(14) http://www.nih.go.jp/mirror/Kinases/pkr/pk_catalytic/pk_hanks_se-
q_align_long.html.
(15) Atherton-fessler S.; Parker L. L.; Geahlen R. L.; Piwnica-worms
H. Mechanisms of p38cdc2 Regulation. Mol. Cell. Biol. 1993, 13,
1675-1685.
Refer en ces
(1) Pines, J . Cyclins and cyclin-dependent kinases: take your
partners. Trends Biochem. Sci. 1993, 18, 195-197.
(2) Sherr, C. J . D-type cyclins. Trends Biochem. Sci. 1995, 20, 187-
190.
(16) (a) Russo, A. A.; Tong, L.; Lee, J .-O.; J effrey, P. D.; Pavletich,
N. P. Structural basis for inhibition of the cyclin-dependent
kinase Cdk6 by the tumor suppressor p16INK4a. Nature 1998,
395, 237-243. (b) J effrey, P. D.; Tong, L.; Pavletich, N. P.
Structural basis of inhibition of CDK-cyclin complexes by INK4
inhibitors. Genes Dev. 2000, 14, 3115-3125. (c) Brotherton, D.
H.; Dhanaraj, V.; Wick, S.; Leonardo, B.; Domaille, P. J .;
Volyanik, E.; Xu, X.; Parisini, E.; Smith, B. O.; Archher, S. J .;
Serrano, M.; Brenner, S. L.; Blundell, T. L.; Laue, E. D. Crystal
structure of the complex of the cyclin D-dependent kinase Cdk6
bound to the cell-cycle inhibitor p19INK4d. Nature 1998, 395,
244-250.
(17) (a) Lukas, J .; Parry, D.; Aagaard, L.; Mann, D. J .; Bartkova, J .;
Stauss, M.; Peters, G.; Bartek, J . Retinoblastoma-protein-
dependent cell-cycle inhibition by the tumor suppressor p16.
Nature 1995, 375, 503-506. (b) Koh, J .; Enders, G. H.; Dynlacht,
B. D.; Harlow, E. Tumor-derived p16 encoding proteins defective
in cell-cycle inhibition. Nature 1995, 375, 506-508.
(18) Reviewed in Murcko M. A. An introduction to de novo ligand
design. In Practical Application of Computer-aided Drug Design;
Charifson, P. S., Ed.; Marcel Dekker: New York, 1997; pp 304-
354.
(19) (a) Boehm, H. J . The computer program LUDI: a new method
for the de novo design of enzyme inhibitors. J . Comput.-Aided
Mol. Des. 1992, 6, 61-78. (b) Boehm H. J . LUDI: rule-based
automatic design of new substituents for enzyme inhibitor leads.
J . Comput.-Aided Mol. Des. 1992, 6, 593-606.
(3) (a) Taya, Y. Cell cycle-dependent phosphorylation of the tumor
suppressor RB protein. Mol. Cells 1995, 5, 191-195. (b) Wein-
berg, R. A. The retinoblastoma protein and a cell cycle control.
Cell 1995, 81, 323-330. (c) Kato, J .; Matsushime, H.; Hiebert,
S. W.; Ewen, M. E.; Sherr, C. J . Direct binding of cyclin D to
the retinoblastoma gene product (pRB) and pRB phosphorylation
by the cyclin D-dependent kinase CDK4. Genes Dev. 1993, 7,
331-342. (d) Baldin, V.; Lukas, J .; Marcote, M. J .; Pagano, M.;
Draetta, G. Cyclin D1 is nuclear protein required for cell cycle
progression in G1. Genes Dev. 1993, 7, 812-821.
(4) Kamb, A. Cell-cycle regulators and cancer. Trends Genet. 1995,
11, 136-140.
(5) Fry, D. W.; Garret, M. D. Inhibitors of cyclin-dependent kinases
as therapeutic agents for the treatment of cancer. Curr. Opin.
Oncol. Endocr. Metab. Invest. Drugs 2000, 2, 40-59.
(6) Honma, T.; Hayashi, K.; Aoyama, T.; Hashimoto, N.; Machida,
T.; Fukasawa, K.; Iwama, T.; Ikeura, C.; Ikuta, M.; Suzuki-
Takahashi, I.; Iwasawa, Y.; Hayama, T.; Nishimura, S.; Mor-
ishima, H. Structure-based generation of a new class of potent
Cdk4 inhibitors: new de novo design strategy and library design.
J . Med. Chem. 2001, 44, 4615-4627.
(7) Russo, A. A.; J effrey, P. D.; Pavletich, N. P. Structural basis of
cyclin-dependent kinase activation by phosphorylation. Nat.
Struct. Biol. 1996, 3, 696-700.
(8) Iwama, T.; Honma, T.; Ikeura, C. System for evaluation of
availability of essential structures generated by de novo design
programs (SEEDS). Unpublished results.
(9) Ikuta, M.; Kamata, K.; Fukasawa, K.; Honma, T.; Machida, T.;
Hirai, H.; Suzuki-Takahashi, I.; Hayama, T.; Nishimura, S.
Crystallographic approach to identification of cyclin dependent
kinase 4 (CDK4) specific inhibitors by using CDK4 mimic CDK2
protein. J . Biol. Chem. 2001, 276, 27548-27554.
(10) (a) Nurse, P. Universal control mechanism regulating onset of
M-phase. Nature 1990, 344, 503-508. (b) J ohnson, D. G.;
Walker, C. L. Cyclins and cell cycle check-points. Annu. Rev.
Pharmacol. Toxicol. 1999, 39, 295-312.
(20) LeapFrog manual, SYBYL version 6.1, Tripos Associates, St.
Louis, MO.
(21) Elnagdi, M. H.; Abdel-Galil, F. M.; Riad, B. Y.; Elgemeie, G. E.
H. Recent developments in the chemistry of 3(5)-aminopyrazoles.
Heterocycles 1983, 20, 2437-2470.
(22) The absolute configuration of 14a was confirmed by synthesis
of 14a from (-)-N-benzyl-L-proline ethyl ester and (+)-(9bR)-9-
amino-1,2,3,9b-tetrahydro-5H-pyrrolo[2,1-a]isoindol-5-one.6 The
absolute configuration of 15a was confirmed by synthesis of 15a
from (+)-(9bR)-9-amino-1,2,3,9b-tetrahydro-5H-pyrrolo[2,1-a]-
isoindol-5-one6 and (+)-(2S)-5-chloro-2,3-dihydro-1H-inden-2-
amine. The absolute configuration of (-)-(2R)-5-chloro-2,3-
dihydro-1H-inden-2-amine was determined by X-ray analysis.
(23) Hirai, H.; Fukasawa, K.; Machida, T.; Shimomura, T.; Oki, H.;
Honma, T.; Hayashi, K.; Yoshizumi, T.; Hayama, T.; Suzuki-
Takahashi, I. Effects of diarylurea class of Cdk4 selective
inhibitor on the progression of mammalian cell cycle. Unpub-
lished results.
(24) Ege, G.; Arnord, P. Aminopyrazole; II. C-Unsubstituierte 1-alkyl-
3-aminopyrazole aus 2-chloroacrylnitril bzw. 2,3-dichloropro-
pannitril und alkylhydrazinen. Synthesis 1976, 1, 52-53.
(25) Hammond, K. M.; Fisher, N.; Morgan, E. N.; Tanner, E. M.;
Franklin, C. S. 3:5-Dioxo-1:2-diphenylpyrazolidines. The 4-hy-
droxy- and certain 4-alkoxy- and 4-alkylamino-analogues. J .
Chem. Soc. 1957, 1062-1067.
(11) (a) Pardee, A. B. A restriction point for control of normal animal
cell proliferation. Proc. Natl. Acad. Sci. U.S.A. 1974, 71, 1286-
1290. (b) Pardee, A. B.; Durbrow, R.; Hamlin, J . L.; Kletzien, R.
F. Animal cell cycle. Annu. Rev. Biochem. 1978, 47, 715-750.
(12) (a) Chen, J . M.; Nelson, F. C.; Levin, J . I.; Mobilio, D.; Moy, F.
J .; Nilakantan, R.; Zask, A.; Powers, R. Structure-based design
of a novel, potent, and selective inhibitor for MMP-13 utilizing
NMR spectroscopy and computer-aided molecular design. J . Am.
Chem. Soc. 2000, 122, 9648-9654. (b) Iversen, L. F.; Andersen,
H. S.; Branner, S.; Mortensen, S. B.; Peters, G. H.; Norris, K.;
Olsen, O. H.; J eppesen, C. B.; Lundt, B. F.; Ripka, W.; Moller,
K. B.; Moller, N. P. H. Structure-based design of a low molecular
weight, nonphosphorus, nonpeptide, and highly selective inhibi-
tor of protein-tyrosine phosphatase 1B. J . Biol. Chem. 2000, 275,
10300-10307. (c) Bayly, C. I.; Black, W. C.; Leger, S.; Ouimet,
N.; Ouellet, M.; Percival, M. D. Structure-based design of COX-2
selectivity into flurbiprofen. Bioorg. Med. Chem. Lett. 1999, 9,
307-312. (d) Kurumbail, R. G.; Stevens, A. M.; Gierse, J . K.;
McDonald, J . J .; Stegeman, R. A.; Pak, J . Y.; Gildehaus D.;
Miyashiro, J . M.; Penning, T. D.; Seibert, K.; Isakson, P. C.;
Stallings, W. C.; Structural basis for selective inhibition of
cyclooxygenase-2 by antiinflammatory agents. Nature 1996, 384,
644-648.
(13) (a) Hanks, S. K.; Quinn, A. M.; Hunter, T. The protein kinase
family: conserved features and deduced phylogeny of the
catalytic domains. Science 1988, 241, 42-52. (b) Hanks, S. K.;
Quinn, A. M. Protein kinase catalytic domain sequence data-
base: identification of conserved features of primary structure
and classification of family members. Methods Enzymol. 1991,
200, 38-62. (c) Hanks, S. K.; Hunter, T. Protein kinases. 6. The
eukaryotic protein kinase superfamily: kinase (catalytic) domain
structure and classification. FASEB J . 1995, 9, 576-96. (d) Also
see a review: Toledo, L. M.; Lydon, N. B.; Elbaum, D.; The
structure-based design of ATP-site directed protein kinase
inhibitors. Curr. Med. Chem. 1999, 6, 775-805.
(26) (a) Kitagawa, M.; Okabe, T.; Ogino, H.; Matsumoto, H.; Suzuki-
Takahashi, I.; Kokubo, T.; Higashi, H.; Saitoh, S.; Taya, Y.;
Yasuda, H.; Ohba, Y.; Nishimura, S.; Tanaka, N.; Okuyama, A.
Butyrolactone I, a selective inhibitor of cdk2 and cdc2 kinase.
Oncogene 1993, 8, 2425-2432. (b) Kitagawa, M.; Higashi, H.;
J ung, H.-K.; Suzuki-Takahashi, I.; Ikeda, M.; Tamai, K.; Kato,
J .; Segawa, K.; Yoshida, E.; Nishimura, S.; Taya, Y. The
consensus motif for phosphorylation by cyclin D1-Cdk4 is
different from that for phosphorylation by cyclin A/E-Cdk2.
EMBO J . 1996, 15, 7060-7069.
(27) MDS Pharma Services Discovery, WA.
(28) Park, Y.-W.; Cummings, R. T.; Wu, L.; Zheng, S.; Cameron, P.
M.; Woods, A.; Zaller, D. M.; Marcy, A. I.; Hermes, J . D.
Homogeneous proximity tyrosine kinase assays: scintillation
proximity assay versus homogeneous time-resolved fluorescence.
Anal. Biochem. 1999, 269, 94-104.
J M010326Y