Structure-based design of (5-arylamino-2H-pyrazol-3-yl)-biphenyl-2′, 4′-diols as novel and potent human CHK1 inhibitors

Article abstract of DOI:10.1021/jm0704604

The cocrystal structure of a library hit was used to design a novel series of CHK1 inhibitors. The new series retained the critical hydrogen-bonding groups of the resorcinol moiety for binding but lacked the phenolic anilide moiety. The newly designed com

Full text of DOI:10.1021/jm0704604

J. Med. Chem. 2007, 50, 5253-5256
5253
Structure-Based Design of
(5-Arylamino-2H-pyrazol-3-yl)-biphenyl-2′,4′-diols
as Novel and Potent Human CHK1 Inhibitors
Min Teng,*,† Jinjiang Zhu,† Michael D. Johnson,†
Ping Chen,‡ Jill Kornmann,§ Enhong Chen,§
Alessandra Blasina,§ James Register,| Kenna Anderes,§
Caroline Rogers,‡ Yali Deng,‡ Sacha Ninkovic,†
Stephan Grant,| Qiyue Hu, Karen Lundgren,§,#
Zhengwei Peng, and Robert S. Kania†
Figure 1. Our library hits: indazole CHK1 inhibitors.
of G2 checkpoint by modulating Cdc25, a dual specificity
phosphatase that activates Cdc2. In eukaryotes, Cdc2 is also
known as cyclin-dependent kinase 1 (CDK1), which promotes
mitotic entry.5 CHK1 transduces signals from the DNA dam-
age sensory complex to inhibit activation of Cdc2-cyclin B
complex. CHK1 serves as the direct link between the G2
checkpoint and the negative regulation of Cdc2. Inactivation
of CHK1 has shown to abrogate G2 arrest induced by DNA
damage. When administered during the course of a DNA
damaging event, such as chemotherapy employing antineoplastic
agents, radiation therapy, immunotherapies, or antiangiogenic
therapies, a CHK1 inhibitor should sensitize cancer cells, thereby
triggering damage-mediated apoptosis. Meanwhile, healthy cells
should retain G1 checkpoint regulated by p53 to avoid death.6
To not exacerbate the toxicities associated with chemo- and
radiotherapies, an ideal CHK1 inhibitor should display no
activity on its own. Administered as adjuvant therapeutics, a
selective CHK1 inhibitor could prove beneficial for the treatment
of cancer.7
There have been significant interests in the identification of
CHK1 inhibitors.8 Among the more well-known inhibitors is
Staurosporine, a natural product whose analogs include an
indolocarbazole (SB218078, SmithKline Beecham)9 and 7-hy-
droxystaurosporine (UCN-01, NCI/Kyowa-Hakko),10 the latter
currently being evaluated in clinical trials as an anticancer
agent.11 These compounds display great inhibitory potency
against both CHK1 and CDK1.9 A number of other CHK1
inhibitors have been reported;12 however, there is a clear need
for potent and selective CHK1 inhibitors derived from a variety
of chemotypes. Previously, our labs solved the structure of the
human CHK1 kinase domain (residues 1-289) and its binary
complex with AMP-PNP (1.7 Å), which facilitated the design
of novel chemical classes.13 Herein, we report the results of
optimization of a screening hit using a structure-based drug
design (SBDD) approach.
High-throughput screening of targeted libraries against the
CHK1 kinase was employed to identify indazole CHK1 inhibi-
tors represented by 1 and 2 (Figure 1). The library design was
carried out by using an in-house tool called PGVL (Pfizer Global
Virtual Library). Bearing an additional phenolic group, 2 showed
significant improvement over 1 in terms of enzymatic potency
against CHK1 (Table 1). When evaluated in the histone H3
ELISA assay, using camptothecin as the DNA damaging agent,
2 exhibited cellular activity of 550 nM, while 1 displayed no
activity when tested at 1.0 uM and 10 uM.
To elucidate the enzymatic potency difference displayed by
1 and 2, as well as to understand the binding mode of these
indazoles, the X-ray cocrystal structures (1.7 Å) of 1 and 2
bound to the CHK1 enzyme were obtained. Both indazoles bind
within the ATP binding pocket as anticipated (Figure 2a,b). For
Department of Medicinal Chemistry, Biochemical Pharmacology,
Research Pharmacology, Crystallography, and Computational
Chemistry, Pfizer Global Research and DeVelopment, La Jolla, Inc.,
10770 Science Center DriVe, San Diego, California 92121-1194
ReceiVed April 18, 2007
Abstract: The cocrystal structure of a library hit was used to design
a novel series of CHK1 inhibitors. The new series retained the critical
hydrogen-bonding groups of the resorcinol moiety for binding but
lacked the phenolic anilide moiety. The newly designed compounds
exhibited similar enzymatic activity, while demonstrating increased
cellular potency. Compound 10c, showing no single agent effect,
potentiated the antiproliferative effect of Gemcitabine in both prostate
and breast cancer cell lines.
A eukaryotic cell cycle has a carefully regulated progression
of phases: initial gap (G1), DNA synthesis (S), secondary gap
(G2), and mitosis (M). Together, G1, S, and G2 are known as
interphases. In G1, the cell, whose biosynthetic pathways were
slowed during mitosis, resumes a high rate of RNA and protein
biosynthesis. The S phase begins when DNA synthesis starts
and ends when the DNA content of the nucleus has been repli-
cated. The cell then enters G2, where RNA and protein biosyn-
thesis occur once more. Following G2, the cell enters M phase,
which begins with nuclear division and ends with the complete
division of the cytoplasm into two daughter cells. This marks
the beginning of the interphase for new cells. Nondividing cells
exist at G0, a time following mitosis and before DNA synthesis.
Cells going into mitosis with damaged DNA will suffer from
mitotic catastrophe, which ultimately leads to apoptosis. There-
fore, in response to DNA damages, cells undergo various cell
cycle arrests (G1, S, G2) to allow time for DNA repair.1 Healthy
cells have both G1 and G2 checkpoints.2 The repair processes
associated with G1 and G2 help to ensure cell viability after
DNA damage by chemotherapy and radiation. Many types of
cancer cells, however, rely exclusively on the G2 checkpoint
and its associated repair processes to remain viable and continue
replication.3 Abrogation of the G2 checkpoint would leave cancer
cells with no means to delay progression into mitosis following
DNA damage.
Checkpoint enzymes, such as the serine/threonine protein
kinase called checkpoint kinase 1 (CHK1a1 or p56CHK1), are
responsible for maintaining the order and fidelity of events in
the cell cycle.4 CHK1 plays an essential role in the regulation
* To whom correspondence should be addressed. Tel.: (858) 401-8254.
Fax: (858) 795-9333. E-mail: min.teng@biogenidec.com.
† Department of Medicinal Chemistry.
‡ Department of Crystallography.
§ Department of Research Pharmacology.
| Department of Biochemical Pharmacology.
Department of Computational Chemistry.
# Current address: 5200 Research Place, Biogen Idec, San Diego, CA
92122.
a Abbreviations: CHK1, checkpoint kinase 1; SBDD, structure-based
drug design; CDK1, cyclin-dependent kinase 1; ATP, adenosine triphos-
phate; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide;
FACS, fluorescence activated cell sorting.
10.1021/jm0704604 CCC: $37.00 © 2007 American Chemical Society
Published on Web 09/21/2007

5254 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 22
Table 1. Biological Potency of Compounds 1-3, 10a-g at CHK1
Letters
cmpd
X
R
Kia (nM)
EC50b (nM)
Figure 3. Compounds 3 (aqua) and 2 (gold) as bound in the ATP
binding site of CHK1.
1
2
3
10a
10b
10c
10d
10e
10f
10g
301
0.5
NAc
550
500
70
41
60
1.20
0.36
0.18
0.38
0.19
0.30
0.20
0.80
CH
CH
CH
CH
N
iso-propylamino
pyrrolidinyl
cyclopropylamino
dimethylamino
piperidinyl
as shown in Figure 2b. As a result, the phenolic anilide portion
of the two inhibitors point in different directions (Figure 2c).
This change of the amide conformation in 2 enables the
2-hydroxy to form an internal H-bond with the amide NH of 2
(2.30 Å) and a H-bond with w-749 (2.62 Å), which in turn forms
H-bonds with Val68 (2.77 Å) and Asn59 (2.82 Å; Figure 2b).
The 4-hydroxy group is in place to form a H-bond directly with
Asn59 (2.93 Å), extruding w-690 in addition to forming a
H-bond with Glu55 (2.43 Å). The sophisticated H-bond network
involving w-749, Val68, Asn 59, Glu55, and the dihydroxy from
the resorcinol moiety appears to be responsible for the superb
potency of 2 against CHK1.
58
150
230
165
N
N
sec-butylamino
cyclopropylmethylamino
a Data is the mean of duplicates with a SD of (5%. b Data is the mean
of duplicates with a SD of (5.53%. c NA means no activity when tested at
1.0 uM and 10 uM.
Although 2 showed excellent enzymatic potency against
CHK1, its phenolic anilide feature was a concern due to the
potential for it to form a reactive metabolite.14 In addition,
cellular potency required further improvement. Information
gained from the cocrystal structures (Figure 2) indicated that,
unlike 1, the amide linker in 2 offered no interaction with the
backbone of the CHK1 ATP pocket. The amide does, however,
project the resorcinol head group to enable the desired hydrogen
bond network. Moreover, in the environment where the amide
linker resides there was additional unoccupied pocket space. A
closer examination of this pocket revealed that it is lined mainly
by lipophilic residues from Leu84 and Val68. This information
inspired designs that replaced the amide linker in 2 with more
liphophilic linkers that rigidly maintained proper distance and
vectors to present the head group. These designs also targeted
cellular potency optimization for which we aimed to improve
permeability as a key strategy. In addition to amide replacement,
we looked to reduce the desolvation penalty associated with
the benzimidazole moiety. We replaced the benzimidazole with
an amino pyridyl or amino phenyl group since these groups
maintain the H-bond donating ability to the backbone carbonyl
of Cys87 while possessing less polar surface areas.
The structural transformations used to accomplish these
design principles evolved over time and are summarized as
follows: First, the benzimidazole ring is changed into an aniline
moiety as shown in A (Scheme 1). The next modification
involves the truncation of the indazole ring into a pyrazole ring,
as shown in B, preserving the acceptor-donor interactions with
Cys87 and Glu85 of the enzyme. A new six-member ring is
then constructed in the process of which the amide linker
becomes an integral part of the ring, as in C. Conversion of the
pyridone C into a phenyl ring leading to 3 completes the entire
structural transformations. As a result, a phenyl ring in 3 is used
as a rigid linker to replace the amide in 2.
Figure 2. (a,b) Cocrystal structures of 1 and 2 bound to the ATP
binding site of CHK1. To show H-bond interactions, hydrogen atoms
are added for all cocrystal structures. (c) Overlay of X-ray structures
of indazole 1 and 2 in the ATP binding pocket of CHK1.
both compounds, the NH of the benzimidazole, together with
the H-bound donor and acceptor of the indazole, form two
H-bonds with Cys87 (backbone carbonyl and NH) and one with
Glu85 (backbone carbonyl) of the hinge region of the CHK1
ATP binding pocket. Binding differences between 1 and 2 are
derived from the differing anilide head groups. The para-
hydroxy in 1 forms a H-bond with Glu55. The amide linker in
1 interacts with Val68 and Asn59 through w-694 and w-693
(Figure 2a). For 2, the amide linker flips 180° to extrude w-694,
Compounds 3, 10a-g were synthesized according to previ-
ously described procedures.15
As expected, the cocrystal structure of 3 with the CHK1
enzyme (1.85 Å) was essentially identical to that of 2 with
respect to the resorcinol region (Figure 3). This part of the
molecule is clearly central to the binding of these inhibitors.
The center phenyl ring now fills in the lipophilic “space” to
create van der Waals interactions with the residue of Leu84.

Letters
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 22 5255
Scheme 1. Structural Transformation from Compound 2 to Compound 3 Guided by SBDD
structural modification at the para-position of the aniline could
provide opportunities for further potency improvement. We
decided to introduce an alkyl amino group to the para-position
of the aniline or pyridyl ring. It was anticipated that this alkyl
amino group would provide additional hydrogen bond interac-
tions with Glu91 and Asp94 without significantly increasing
the polar surface area of the molecule.
Table 1 lists a number of compounds with such a modification
and their enzymatic and cellular potency against CHK1. As
shown, the addition of a simple alkyl amino group to the para-
position of the aniline or pyridyl amine improved both enzy-
matic and cellular activity compared to that of 3. Compound
10a was evaluated against a panel of 36 kinases. The selec-
tivity data indicated that 10a displayed good selectivity as its
activity at CHK1 is at least 85-fold more potent than at other
kinases.
Compound 10c was evaluated for the potentiation of the
growth inhibitory activity of Gemcitabine in prostate and breast
cancer cell lines in the MTT assay. In this assay, increasing
concentrations of Gemcitabine were used in combination with
240 nM of 10c. When used alone, 10c induced no or minimal
cellular toxicity (Figure 4). The combination treatment resulted
in greater activity (IC50 of 1.0 nM in prostate, 3.8 nM in breast
cancer cells) than the treatment by Gemcitabine alone (IC50 of
3.5 nM in prostate, 11.4 nM in breast cancer cells). These results
provided evidence for the in vitro efficacy of 10c in enhancing
the cancer cell killing activity of Gemcitabine with minimal
single agent activity attributed to 10c alone.
Figure 4. Potentiation of the antiproliferative effect of Gemcitabine
by 10c at 0.24 uM.
The H-bond interactions between the pyazole nitrogens and
Cys87 and Glu85 resemble that made by 2. The NH from the
aniline provides a hydrogen bond with Cys87. The para-position
of the aniline is near the solvent channel with 5.5 Å from Glu91
and 3.9 Å from Asp94.
As seen in Table 1, 3 displayed enzymatic and cellular
potency similar to that of 2 against CHK1. Clearly, the cell
potency needed further improvement. As indicated above, further
To demonstrate that the synergy with Gemcitabine is due to
the inhibition of CHK1 through the abrogation of the S
Figure 5. Fluorescence activated cell sorting (FACS) results of 10a in combination with Gemcitabine.

5256 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 22
Letters
(5) Peng, C.-Y.; Graves, P. R.; Thoma, R. S.; Wu, Z.; Shaw, A. S.;
Piwnica-Worms, H. Mitotic and G2 Checkpoint Control: Regulation
of 14-3-3 Protein Binding by Phosphorylation of Cdc25C on Serine-
216. Science 1997, 277, 1501-1505. (b) Furnari, B.; Rhind, N.;
Russell, P. Cdc25 Mitotic Inducer Targeted by Chk1 DNA Damage
Checkpoint Kinase. Science 1997, 277, 1495-1497.
(6) Nurse, P. Checkpoint Pathways Come of Age. Cell 1997, 91, 865-
867. (b) Weinert, T. Cell Cycle: A DNA Damage Checkpoint Meets
the Cell Cycle Engine. Science 1997, 277, 1450-1451. (c) Walworth,
N.; Davey, S.; Beach, D. Fission Yeast chkl Protein Kinase Links
the rad Checkpoint Pathway to cdc2. Nature 1993, 363, 368-
371.
(7) Certain CHK-1 inhibitors have been proposed for cancer therapy.
(a) Sanchez, Y.; et al. Conservation of the Chk1 Checkpoint Pathway
in Mammals: Linkage of DNA Damage to Cdk Regulation through
Cdc25. Science 1997, 277, 1497-1501. (b) Flaggs, G.; et al. Atm-
Dependent Interactions of a Mammalian Chk1 Homolog with Meiotic
Chromosomes. Curr. Biol. 1997, 7, 977-986.
(8) Prudhomme, M. Novel Checkpoint 1 Inhibitors. Rec. Pat. Anti-Cancer
Drug DiscoVery 2006, 1, 55-68. (b) Tao, Z.-F.; Lin, N.-H. Chk1
Inhibitors for Novel Cancer Treatment. Anti-Cancer Agents Med.
Chem. 2006, 6, 377-388.
(9) Jackson, J. R.; Gilmartin, A.; Imburgia, C.; Winkler, J. D.; Marshall,
L. A.; Roshak, A. An Indolocarbazole Inhibitor of Human Checkpoint
Kinase (Chk1) Abrogates Cell Cycle Arrest Caused by DNA Damage.
Cancer Res. 2000, 60, 566-572.
(10) Graves, P. R.; Yu, L.; Schwarz, J. K.; Gales, J.; Sausville, E. A.; et
al. The CHK1 protein Kinase and the Cdc25C Regulatory Pathways
are Targets of the Anticancer Agent UCN-01. J. Biol. Chem. 2000,
275, 5600-5605. (b) Busby, E. C.; Leistritz, D. F.; Abraham, R. T.;
Karnitz, L. M.; Sarkaria, J. N. The Radiosensitizing Agent 7-Hy-
droxystaurosporine (UCN-01) Inhibits the DNA Damage Checkpoint
Kinase hCHK1. Cancer Res. 2000, 2000, 2108-2112.
(11) Sausville, E. A.; Arbuck, S. G.; Messmann, R.; Headless, D.; Lush,
R. D.; et al. Phase I Trial of 72-Hour Continuous Infusion UCN-01
in Patients with Refractory Neoplasms. J. Clin. Oncol. 2001, 19,
2319-2333. (b) Senderowicz, A. M. The Cell Cycle as a Target for
Cancer Therapy: Basic and Clinical Findings with the Small
Molecule Inhibitors Flavopirodol and UCN-01. Oncologist 2002, 7,
12-19.
checkpoint, fluorescence activated cell sorting (FACS) was
performed with 10a in MDA-MB-231 cell line. As shown in
Figure 5, 10a alone (560 nM) displayed an insignificant effect
on cell cycle, while Gemcitabine alone (10 nM) induced a
prominent S phase arrest at 16 h post-treatment. In the com-
bination treatment, flow cytometry analysis showed abrogation
of Gemcitabine-induced S phase arrest. The time-dependent
decrease in the S phase cells induced by 10a corresponded to
an increase in the G2-M and G0-G1 cell populations, indicating
that the cells were entering mitosis and attempting to re-enter
the cell cycle. In addition, an increase in the population of sub-
G1 cells was detected in the combination-treated cells, indicating
apoptotic cell death.
In conclusion, we have identified novel and potent CHK1
inhibitors through structural transformations of a library hit using
SBDD. Through these exercises, we have illustrated how the
structural information about the target guided us to formulate
and test our ideas. Using this powerful tool, we have removed
the phenolic anilide feature from the original hit and improved
cellular potency. Compound 10c potentiated the growth inhibi-
tory activity of Gemcitabine in both prostate and breast cancer
cell lines. Broader SAR and other biological attributes regarding
the newly identified pyrazole inhibitors will be subjects of future
investigations to be reported in due course.
Acknowledgment. We thank Mr. Hai Wang, Ms. Christine
Thomas, and Mr. Haresh Vazir for hit resynthesis, Dr. Dilip
Bhumaralkar for effort in high-throughput chemistry, Dr. Chun
Luo for protein purification, and Dr. Steve Bender and Dr.
Patrick O’Connor for helpful discussions.
Note Added after ASAP Publication. This manuscript was
released ASAP on September 21, 2007 before final corrections
were incorporated. The corrected version was posted on
September 27, 2007.
(12) Foloppe, N.; Fisher, L. M.; Francis, G.; Howes, R.; Kierstan, P.;
Potter, A. Identification of a Buried Pocket for Potent and Selective
Inhibition of CHK1: Prediction and Verification. Bioorg. Med. Chem.
2006, 14, 1792-1804 and references cited therein. (b) Tao, Z.-F.; et
al. Structure-Based Design, Synthesis, and Biological Evaluation of
Potent and Selective Macrocyclic Checkpoint Kinase 1 Inhibitors.
J. Med. Chem. 2007, 50 (7), 1514-1527. (c) Tong, Y.; et al.
Discovery of 1,4-Dihydroindeno[1,2-c]pyrazoles as a Novel Class
of Potent and Selective Checkpoint Kinase 1 Inhibitors. Bioorg. Med.
Chem. 2007, 15, 2759-2767.
(13) Chen, P.; Luo, C.; Deng, Y.; Ryan, K.; Register, J.; Margosiak, S.
Tempczyk-Russell, A.; Nguyen, B.; Myers, P.; Lundgren, K.; Kan,
C.-C.; O’Connor, P. M. The 1.7 Å Crystal Structure of Human Cell
Cycle Checkpoint Kinase Chk1: Implications for Chk1 Regulation.
Cell 2000, 100, 681-692.
(14) Benya, T. J.; Cornish, H. H. Toxicology of Aromatic Nitro and Amino
Compounds. Patty’s Industrial Hygiene and Toxicology (4th Edition).
1994, 2 (Pt. B), 947-1085. (b) Kazius, J.; McGuire, R.; Bursi, R.
Derivation and Validation of Toxicophores for Mutagenicity Predic-
tion. J. Med. Chem. 2005, 48 (1), 312.
(15) Johnson, M. D.; Teng, M.; Zhu, J. Preparation of aminopyrazoles as
CHK1 checkpoint protein kinase inhibitors. U.S. patent application,
WO 2005009435, 2005.
Supporting Information Available: Synthesis procedures and
characterization data for intermediates and final compounds.
Procedures for enzymatic, cellular, MTT, and FACS assays.
Crystallization conditions and X-ray data statistics for 1, 2, and 3.
Kinase selectivity profile for 10a. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Zhou, B.-B.; Elledge, S. J. The DNA Damage Response: Putting
Checkpoints in Perspective. Nature 2000, 408, 433-439. (b) Melo,
J.; Toczyski, D. A Unified View of the DNA-Damage Checkpoint.
Curr. Opin. Cell Biol. 2002, 14, 237-245.
(2) Li, Q.; Zhu, G.-D. Targeting Serine/Threonine Protein Kinase B/Akt
and Cell-Cyclo Checkpoint Kinases for Treating Cancer. Curr. Top.
Med. Chem. 2002, 2, 939-971.
(3) Kastan, M. B.; Onyekwere, O.; Sidransky, D.; Vogelstein, B.; Craig,
R. W. Participation of p53 Protein in the Cellular Response to DNA
Damage. Cancer Res. 1991, 51, 6304-6311. (b) Greenblatt, M. S.;
Bennett, W. P.; Hollstein, M.; Harris, C. C. Mutations in the p53
Tumor Suppressor Gene: Clues to Cancer Etiology and Molecular
Pathogenesis. Cancer Res. 1994, 54, 4855-4878..
(4) Sancar, A.; Lindsey-Boltz, L. A.; Unsal-Kacmaz, K.; Linn, S.
Molecular Mechanisms of Mammalian DNA Repair and the DNA
Damage Checkpoints. Annu. ReV. Biochem. 2004, 73, 39-85. (b)
Nasmyth, K. Putting the Cell Cycle in Order. Science 1996, 274,
1643.
JM0704604