Anilinopyrazole as selective CDK2 inhibitors: Design, synthesis, biological evaluation, and X-ray crystallographic analysis

Article abstract of DOI:10.1016/S0960-894X(03)00630-9

A novel series of anilinopyrazoles has been designed based on the X-ray crystal structure analysis. Most compounds from this series not only show sub-nanomolar IC₅₀ values for CDK2, but also demonstrate almost 1000-fold selectivity to other kinases including CDK1.

Full text of DOI:10.1016/S0960-894X(03)00630-9

Bioorganic & Medicinal Chemistry Letters 13 (2003) 2985–2988
Anilinopyrazole as Selective CDK2 Inhibitors:
Design, Synthesis, Biological Evaluation, and X-ray
Crystallographic Analysis
Jun Tang,a,* Lisa M. Shewchuk,b Hideyuki Sato,a Masaichi Hasegawa,a
Yoshiaki Washioa and Naohiko Nishigakia
aGlaxoSmithKline K. K. Tsukuba Research Laboratories, 43 Wadai Tsukuba, Ibaraki 300-4247, Japan
bGlaxoSmithKline Inc., Five Moore Drive, Research Triangle Park, NC 27709, USA
Received 12 February 2003; accepted 14 March 2003
Abstract—A novel series of anilinopyrazoles has been designed based on the X-ray crystal structure analysis. Most compounds
from this series not only show sub-nanomolar IC50 values for CDK2, but also demonstrate almost 1000-fold selectivity to other
kinases including CDK1.
# 2003 Elsevier Ltd. All rights reserved.
Total 518 protein kinases identified from the human
genome sequence and 130 protein phosphatases exert
precise and reversible control on protein phosphoryla-
tion.1 Through this function, they play pivotal roles in
regulating aspects of metabolism, gene expression, cell
growth, cell division and cell differentiation. The reg-
ulation of the cell cycle is a well orchestrated pairing of
different family members of the Cyclin Dependent
Kinases (CDKs) and cyclin regulatory subunits.2,3 The
inhibition of CDK2should arrest cells in G1 phase and
prevent them from entering the cell cycle. Therefore,
inhibitors of CDK2may have utility in the treatment of
proliferative diseases such as cancer and psoriasis.4
When dealing with protein kinases, one major concern
is selectivity among the different kinases due to the
markedly conserved structure of the catalytic domain.
Regarding CDKs, structural information indicates that
the catalytic center includes a core of ca. 300 amino
acids which possesses a high degree of homology
between the different members of this protein kinase
family.5 CDK inhibitors acting as ATP competitive
ligands have been thoroughly reviewed in the literature.3
Achieving selectivity between CDK1 and CDK2
remains a challenge since these enzymes share 64%
identity overall and 90% identity in ATP binding pocket.6
Generation of a highly selective CDK2inhibitor series
emerged from our kinase system based research.7
crystal structure of 1a (where R=m-Br), composed of
cyclin A/CDK2, was successfully obtained with good
resolution and this co-crystallized structure allowed a
good understanding of the necessary selectivity require-
ments. This communication describes our efforts in
exploring a selective CDK2inhibitor, represented by
structure 1, which not only shows excellent selectivity
against other kinases such as VEGFR2, SRC, and
GSK3 but also shows excellent selectivity against CDK1
(Scheme 1).
A
The initial objective of our program was to generate
diverse kinase inhibitors using a system-based-research
strategy,7 from which a CDK2hit compound 1a, was
identified with good activity and selectivity. To this end,
synthetic methods to explore the activity and selectivity
*Corresponding author. Tel.: +81-29-864-5050; fax: +81-29-864-
8558; e-mail: jt21064@gsk.com
Scheme 1. Anilinopyrazole 1.
0960-894X/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00630-9

2986
J. Tang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2985–2988
Scheme 3. Reagents: (a) 4-aminobenzene sulfonamide, toluene,
140 ꢀC, 3 h; (b) hydrazine monohydrate, HOAc, EtOH, 85 ꢀC, 12h.
Scheme 2. Reagents: (a) NaH, carbonic acid dimethyl ester (excess),
75 ꢀC, 20 min; (b) 4-aminobenzene sulfonamide, toluene/DMF=3/1,
130 ꢀC, 12h; (c) Lawesson’s reagent, toluene, 140 ꢀC, 1 h; (d) hydrazine
monohydrate, HOAc, EtOH, 85 ꢀC, 12h.
with this method. An alternative procedure as shown in
Scheme 4 involved the reaction of acetophenone with an
isothiocyanate, followed by the same cyclization
approach with hydrazine to afford the corresponding
desired compounds 1c–1g and 1i–1k, which are exem-
plified in Table 1. Further transformations of the above
compounds can be carried out to explore other function-
alization. For example, anilinopyrazole sulfonamide
derivative 1o can be prepared via a reduction of the
nitro- intermediate with Na2S, followed by a reaction
with benzenesulfonyl chloride as indicated in Scheme 4.8
of the anilinopyrazoles have been established and are
summarized in Schemes 2–4.
Preparation of the anilinopyrazole compounds was
generally achieved by the reaction of substituted acet-
ophenone with carbonic acid dimethyl ester, followed
by treatment with aniline (Scheme 2). The correspond-
ing thiopropionamide was obtained by reaction with
Lawesson’s reagent, followed by treatment with hydra-
zine to give the desired analogues. This protocol was
used to provide compounds 1a and, 2 as indicated in
Table 1. Commercially available 1-(4-fluoro-phenyl)-
3,3-dimercaptopropenone could also be used as starting
materials for the preparation of anilinopyrazole analo-
gues (Scheme 3). Compounds 1b and, 3 were obtained
The crystal structure of CDK2complexed with 1a was
determined with high resolution. The inhibitor bound
into the ATP binding pocket is shown in Figure 1. Two
hydrogen bonds were formed between the aminopyr-
azole moiety of compound 1a and CDK2. Specifically,
the sulfonamide NH and amide carbonyl oxygen were
Table 1. CDK1 and CDK2inhibitory activities of 1a–1o, 2, and 3
Compd
R1
R2R3
R4
Kinase IC
50 (nM)
CDK2
CDK1
1a
2
3
H
–SO2NH2
–NHSO2CH3
H
H
H
F
F
Br
Br
Br
H
H
H
H
F
F
H
H
ND
ND
ND
ND
126
ND
ND
200
ND
501
ND
398
398
ND
501
794
501
23.4
602
871
17.4
H
–SO2NH2
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1o
H
H
H
H
H
H
H
H
H
H
H
H
H
H
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
10
>2000
407
–N(CH3)2
–OCH3
–OH
–NO2
–NH2
H
12.6
0.33
0.34
490
24.5
9.1
166
10.7
61.6
10.7
–Morpholine
–OMe
–OH
–NH2
–NHSO2CH3
H
H
H
H
H
H
–NHCONH–(2-fluoro-5-trifluorophenyl)
–NHSO2-phenyl

J. Tang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2985–2988
2987
both H-bonded with Asp86. The para-sulfonamide on
the 3-anilino group proved to be an essential function-
ality, based on the fact that all other substitutions dra-
matically decreased the inhibitory potency. For
replaces sulfonamide demonstrated a 25-fold reduction
in inhibitory activity. Furthermore, attaching sulfona-
mide at the meta-position (Table 1, compound 3) dra-
matically reduced the activity. In the ligand bound
crystal structure, the bromophenyl ring of compound 1a
occupies a pocket near Lys33 composed of Lys33,
Glu51, Asp145, Val64 and Phe80. This area including
Phe80, is smaller in size, relative to many other kinases
such as GSK3 and VEGFR2, and confers a very
important opportunity for selectivity. We propose that
if this space could be perfectly filled by a substituent,
this will result in not only high activity but also excellent
selectivity. The 10- and 20-positions at the backside of
the bromophenyl ring reside near the surface of
Val64 and Phe80 suggesting that there is no space for
accommodating substituents of any size. The 30-sub-
stituent on the phenyl ring would occupy a small
cavity close to Glu51 and a large substituent would be
expected to encounter unfavorable steric interactions
with the enzyme in this position. The 40-position pro-
jects toward a cleft composed by Lys33 and Asp145,
and into the region near the sugar pocket. This struc-
tural characteristic suggested a large substituent could
be tolerated.
example, compound
2 where methanesulfonamide
The SAR of the anilinopyrazoles proved very consistent
with these structural considerations. Accordingly, sub-
stituents R3 were size limited as evidenced by the inhi-
bitory activities of compounds 1b–1h (Table 1).
Compounds 1g, 1h with a particularly favorable steric
fit for the cavity showed the most potent inhibition of
CDK2with IC 50s in the sub-nanomolar range and
excellent selectivity relative to other kinases including
CDK1. A slightly larger substituent such as the dime-
thylamino group showed >100-fold less potent CDK2
inhibition. Many compounds containing R4 sub-
stituents have also been studied. A morpholine sub-
stituent (compound 1i) was anticipated to interact
unfavorably with the cleft composed by Lys33 and
Asp145, and this substitution dramatically decreased
the activity. Favorable substituents like bromo, meth-
oxy, hydroxy (compounds 1a, 1j, 1k) showed more
potent inhibition with IC50 values less than 25 nM. In
addition, the amino group (compound 1l) was found to
show moderate inhibition of CDK2and derivation to
give sulfonamide and urea substituents provided
potent CDK2inhibitors in contrast to the parent
compound.
Scheme 4. Reagents: (a) 4-isothiocyanato-benzenesulfonamide,
LiN(TMS)2, ꢁ78 ꢀC to rt, THF, 12h; (b) hydrazine monohydrate,
HOAc, EtOH, 85 ꢀC, 12h; (c) Na 2S aq, EtOH, 80 ꢀC, 5 h; (d) benze-
nesulfonyl chloride, pyridine, 0 ꢀC to rt, 12h.
The compounds in the disubstituted pyrazole series
described here were generally about 100-fold more
potent as inhibitors of CDK2than other kinases. For
example, compound 1h bearing an amino group at the
30-position, which was considered to fill the Lys33
pocket best, showed more than 1000-fold selectivity
against other kinases screened [kinases, IC50 (nM)]:
GSK-3, 10,000; SRC, 3981; VEGFR-2, 6300; EGFR,
>10,000; ERBB-4, >10,000; TIE-2, 6300.9
Figure 1. Crystal structure of 1a bound to activated CDK2. Surface
representation of CDK2illustrating the Lysine pocket composed of
Lys33, Glu51, Asp145, Val64 and Phe80. The sulfonamide function-
ality was observed to be positioned at the opening to the ATP binding
cleft. Inhibitor atoms colored as follows: C, gray; N, blue; O, red; S,
yellow; Br, purple.
In conclusion, we have identified a series of novel,
potent CDK2inhibitors. This series also exhibits an
excellent selectivity profile against other kinases
screened to date.

2988
J. Tang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2985–2988
4. (a) Meijer, L.; Leclerc, S.; Leost, M. Pharmacol. Ther. 1999,
Acknowledgements
82, 279. (b) Garrett, M. D.; Fattaey, A. Curr. Opin. Genet.
Dev. 1999, 9, 104. (c) Webster, K. R. Expert Opin. Invest.
Drugs 1998, 7, 865.
5. Structural description of CDK2, please, see: Johnson, L. N.;
Lewis, R. Chem. Rev. 2001, 101, 2209.
6. Toogood, P. L. Med. Res. Rev. 2001, 21, 487, and refer-
ences cited therein.
7. Frye, S. V. Chem. Biol. 1999, 6, R3.
We are grateful to the enzymologists for contributions
to the screening work; Warren Rocque for purifying the
protein for assays and structural work and Annie Has-
sell for growing the crystals and collecting the data;
Karen E. Lackey, Stephen V. Frye, and Drake S.
Eggleston for helpful discussions; Joyce G. Rice for
assistance with manuscript formatting.
8. Analytic data of potent CDK2inhibitors 1a: 1H NMR
(400 MHz, DMSO-d6) ppm 6.44 (s, 1H), 7.06 (s, 2H), 7.42 (m,
3H), 7.55 (d, 1H, J=7.6 Hz), 7.64 (d, 2H, J=8.8 Hz), 7.76 (d,
1H, J=7.6 Hz), 8.00 (t, 1H, J=1.8 Hz), 9.10 (s, 1H), 12.76 (s,
1H). LC/MS: m/z 393, 395 (M+1)+. 1h: 1H NMR (400 MHz,
DMSO-d6) ppm 5.34 (s, 2H), 6.05 (s, 1H), 6.59 (d, 2H,
J=8.6 Hz), 7.03 (s, 2H), 7.38 (d, 2H, J=8.6 Hz), 7.43 (d, 2H,
J=8.8 Hz), 7.61 (d, 2H, J=8.8 Hz), 8.96 (s, 1H), 12.23 (s, 1H).
LC/MS: m/z 330 (M+1)+, 32 8 (Mꢁ1)ꢁ. 1l: 1H NMR
(400 MHz, DMSO-d6) ppm 5.19 (s, 2H), 6.12 (s, 1H), 6.55 (d,
1H, J=7.6 Hz), 6.86 (m, 2H), 7.05 (s, 2H), 7.08 (t, 1H,
J=7.6 Hz), 7.44 (d, 2H, J=8.8 Hz), 7.63 (d, 2H, J=8.8 Hz),
9.02(s, 1H), 12.48 (s, 1H). LC/MS: m/z 330 (M+1)+, 32 8
(Mꢁ1)ꢁ.
References and Notes
1. (a) Blume-Jensen, P.; Hunter, T. Nature 2001, 411, 355. (b)
Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.; Sudar-
sanam, S. Science 2002, 298, 1912.
2. (a) Harper, J. W.; Adams, P. D. Chem. Rev. 2001, 101,
2511. (b) Haesslein, J.-L.; Jullian, N. Curr. Topics Med. Chem.
2002, 2, 1037.
3. Bramson, H. N.; Corona, J.; Davis, S. T.; Dickerson, S. H.;
Edelstein, M.; Frye, S. V.; Gampe, R. T.; Harris, P. A.; Has-
sell, A.; Holmes, W. D.; Hunter, R. N.; Lackey, K. E.; Love-
joy, B.; Luzzio, M. J.; Montana, V.; Rocque, W. J.; Rusnak,
D.; Shewchuk, L. M.; Veal, J. M.; Walker, D. H.; Kuyper, L.
F. J. Med. Chem. 2002, 44, 4339, and references cited therein.
9. The procedures for crystallography and enzymology have
been described in ref 3.