Fragment-based and structure-guided discovery and optimization of Rho kinase inhibitors

Article abstract of DOI:10.1021/jm201289r

Using high concentration biochemical assays and fragment-based screening assisted by structure-guided design, we discovered a novel class of Rho-kinase inhibitors. Compound 18 was equipotent for ROCK1 (IC₅₀ = 650 nM) and ROCK2 (IC₅₀ = 670 nM), whereas compound 24 was more selective for ROCK2 (IC₅₀ = 100 nM) over ROCK1 (IC₅₀ = 1690 nM). The crystal structure of the compound 18-ROCK1 complex revealed that 18 is a type 1 inhibitor that binds the hinge region in the ATP binding site. Compounds 18 and 24 inhibited potently the phosphorylation of the ROCK substrate MLC2 in intact human breast cancer cells.

Full text of DOI:10.1021/jm201289r

Brief Article
pubs.acs.org/jmc
Fragment-Based and Structure-Guided Discovery and Optimization
of Rho Kinase Inhibitors
Rongshi Li,*,† Mathew P. Martin,† Yan Liu,† Binglin Wang,† Ronil A. Patel,† Jin-Yi Zhu,† Nan Sun,†
Roberta Pireddu,† Nicholas J. Lawrence,† Jiannong Li,‡ Eric B. Haura,‡ Shen-Shu Sung,§
Wayne C. Guida,† Ernst Schonbrunn,† and Said M. Sebti†
‡
†Department of Drug Discovery and Department of Experimental Therapeutics, H. Lee Moffitt Cancer Center and Research
Institute, 12902 Magnolia Drive, Tampa, Florida 33612, United States
S
* Supporting Information
ABSTRACT: Using high concentration biochemical assays and fragment-based screening assisted by structure-guided design,
we discovered a novel class of Rho-kinase inhibitors. Compound 18 was equipotent for ROCK1 (IC50 = 650 nM) and ROCK2
(IC50 = 670 nM), whereas compound 24 was more selective for ROCK2 (IC50 = 100 nM) over ROCK1 (IC50 = 1690 nM). The
crystal structure of the compound 18−ROCK1 complex revealed that 18 is a type 1 inhibitor that binds the hinge region in the
ATP binding site. Compounds 18 and 24 inhibited potently the phosphorylation of the ROCK substrate MLC2 in intact human
breast cancer cells.
screening have not been utilized frequently due to the
limitation of method sensitivity and the concern that screening
of fragments at high micromolar concentrations results in
increased false-positive and false-negative rates. At the time we
started our fragment-based discovery of Rho kinase (ROCK)
inhibitors, there was virtually no report of using biochemical
assays for fragment-based screening. However, there have been
several recent reports describing the use of biochemical assays
to screen fragments at high concentrations for hit identification
successfully. These include the discovery of aminoindazole
PDK1 inhibitors,10 novel nonpeptidic inhibitors of β-secretase
(BACE1),11 inhibitors of checkpoint kinase 1,12 and Hsp90
inhibitors.13 In this manuscript, we have successfully used
FBDD approaches and identified potent, selective, and cell-
active inhibitors of ROCK1 and -2, enzymes involved in several
pathological conditions such as cardiovascular diseases,
glaucoma, inflammatory disorders, and cancer (see re-
views14−18).
INTRODUCTION
■
Fragment-based drug discovery (FBDD) is a powerful tool and
a promising strategy in drug discovery. This approach is now an
established paradigm, and success stories of fragment-based
drug design and discovery have been reported for enzyme
inhibitors as well as protein−protein interaction disruptors as
anticancer therapeutics.1 FBDD is based on screening small
numbers (up to several thousands) of compounds to find low-
affinity fragments with Kd values in the high micromolar to
millimolar range. By contrast, conventional high-throughput
screening (HTS) attempts to evaluate as many compounds as
possible (often a million or more) in the hope of finding
relatively more potent hits with Kd values of 10 μM or less. The
advantages of fragment-based screening include accessing larger
chemical space than classical HTS and avoiding steric clash to
the target of interest from the larger molecules in traditional
HTS. Chemical space has been estimated to be greater than
1060 molecules with 30 or fewer heavy (nonhydrogen) atoms.2
However, when stereochemistry is not considered, there are 26
million synthetically accessible molecules with 11 or fewer
heavy atoms composed of only first-row elements (C, N, O,
and F).3 Only a very small fraction of this chemical space can
be explored by HTS. FBDD sets out to identify a starting point
by screening a library of small molecules representing fragments
that cover a larger chemical space of druglike matter. Therefore,
high hit rates, more efficient hits, and the ability to sample
chemical diversity more easily are expected from fragment-
based screening. In addition, compounds derived from
fragments tend to have better physicochemical properties.
The majority of fragment-based screening examples have
relied upon biophysical techniques such as protein nuclear
magnetic resonance (NMR),1 X-ray crystallography,4 surface
plasmon resonance (SPR),5 isothermal titration calorimetry
(ITC),6 target immobilized NMR screening (TINS),7 native
mass spectrometry (NMS),8 and weak affinity chromatography
(WAC).9 Conventional biochemical assays for fragment-based
RESULTS AND DISCUSSION
■
We initiated our studies with a small fragment library
containing seven compounds with pyridyl group as a potential
“hinge” binder to ROCKs. Their molecular masses range from
199 to 293 Da. As shown in Table 1, coupling of para-
aminopyridine with nicotinic acid, picolinic acid, isonicotinic
acid, 4-chloronicotinic acid, and 5-bromonicotinc acid using N-
ethyl-N′-(3-dimethylaminopropyl)carbodiimide (EDC) in N,N-
dimethylformamide (DMF) yielded the corresponding amides
(1−5). A para-fluorophenyl group was installed within
compound 4 and 5 via a Suzuki reaction to give rise to 6
and 7, respectively. This small fragment library was subjected to
high concentration (400 μM) biochemical assays to determine
Received: September 29, 2011
Published: January 24, 2012
© 2012 American Chemical Society
2474
dx.doi.org/10.1021/jm201289r | J. Med. Chem. 2012, 55, 2474−2478

Journal of Medicinal Chemistry
Brief Article
Table 1. Synthesis of Fragments 1−7 and IC50 and LE Values
a
a
compd
MW
ROCK1 (IC50, μM)
LE
ROCK2 (IC50, μM)
LE
b
b
1
2
3
4
5
6
7
199.21
199.21
199.21
233.65
278.10
293.30
293.30
35% inhibition
35% inhibition
b
b
44% inhibition
42% inhibition
75.5 18.9
72.5 6.8
270.0 51.0
>400
0.37
0.35
0.30
55.9 11.3
221.4 25.7
357.2 35.3
>400
0.39
0.31
0.29
<0.2
<0.2
<0.2
<0.2
>400
>400
a
b
Data from triplicate experiments. Compound concentration at 400 μM.
Table 2. Synthesis of Fragments and Compounds 8−19 and IC50 and LE Values
a
a
compd
MW
ROCK1 (IC50, μM)
LE
ROCK2 (IC50, μM)
LE
8
9
133.06
253.26
204.10
252.10
266.12
294.15
310.14
296.13
310.14
314.09
280.13
294.15
181.3 50.7
0.51
119.7 20.3
0.53
b
b
ND
ND
10
11
12
13
14
15
16
17
18
19
59.1 15.2
157.1 34.1
13.9 8.7
21.5 7.3
4.6 2.1
0.38
0.27
0.33
0.27
0.31
0.34
0.33
0.35
0.40
0.35
36.1 5.9
61.8 9.6
5.5 3.1
7.5 1.7
2.3 0.6
1.5 0.4
0.8 0.3
1.1 0.3
0.67 0.12
1.07 0.14
0.40
0.30
0.36
0.32
0.33
0.36
0.36
0.37
0.40
0.37
2.9 0.8
2.6 0.6
2.2 1.0
0.65 0.03
2.46 0.55
a
b
Data from triplicate experiments. Not determined.
their in vitro inhibitory activities against ROCK1 and ROCK2
kinases. While both picolinamide (1) and isonicotinamide (2)
showed only 30−40% inhibition on both ROCK1 and ROCK2
at 400 μM, nicotinamide (3) had IC50 values of 75.5 and 55.9
μM for ROCK1 and ROCK2, with ligand efficiencies (LE ≈
−ΔG/HAC, defined as the free energy of binding divided by
the number of nonhydrogen/heavy atoms19) of 0.37 and 0.39,
respectively. 4-Chlorosubsituted nicotinamide (4) decreased
the inhibitory activity against ROCK2 by 4-fold. 5-Bromosub-
stituted nicotinamide (5) decreased ROCK1 and ROCK2
inhibitory activities by up to 6-fold. Larger substituent (para-
fluorophenyl) at the corresponding positions (6 and 7)
diminished the activity even further (Table 1). These results
suggest that the fragments (4−7) bearing bulkier functional
group in the 4- or 5-position of nicotinamide may interfere with
their binding to the hinge region due to steric clashes. This led
us to evaluate other hinge binders to circumvent this potential
problem.
5-Aminoindazole (8) was chosen as a substitute fragment for
para-aminopyridine. The distances from the amino nitrogen to
1-indazole nitrogen and 2-indazole nitrogen are 5.5 and 5.9 Å,
respectively, whereas the distance from the para-amino
nitrogen to pyridine nitrogen is only 4.2 Å. Fragment 8
showed IC50 values of 181 μM for ROCK1 and 120 μM against
ROCK2 with LE being 0.51 and 0.53, respectively (Table 2).
Although N-(1H-indazole-5-yl)acetamide demonstrated im-
proved potency (65 μM for ROCK1 and 40 μM for
ROCK2), the amides derived from the corresponding aromatic
carboxylic acids did not yield potency-improved compounds
(data not shown). This finding led us to identify an extended
linker. As shown in Table 2, dimethyl/indazole urea 10 had
improved potency with IC50 values of 59 μM for ROCK1 and
36 μM against ROCK2 with LEs being 0.38 and 0.40,
2475
dx.doi.org/10.1021/jm201289r | J. Med. Chem. 2012, 55, 2474−2478

Journal of Medicinal Chemistry
Brief Article
Table 3. Synthesis of Compounds 20−27 and IC50 and LE Values
a
a
compd
MW
ROCK1 (IC50, μM)
LE
ROCK2 (IC50, μM)
LE
22
23
24
25
26
27
303.14
333.15
333.15
317.15
347.16
347.16
1.15 0.41
9.07 2.52
1.69 0.17
2.61 0.41
1.41 0.50
31.01 10.51
0.35
0.28
0.31
0.32
0.31
0.24
0.26 0.07
7.52 1.51
0.10 0.03
19.80 3.29
5.36 0.94
32.92 4.67
0.39
0.28
0.38
0.27
0.28
0.23
a
Data from triplicate experiments.
respectively. However, substitution of one of the methyl groups
by a phenyl group and the other with a hydrogen (compound
11) diminished activity. The addition of a methylene moiety
between the urea nitrogen and the phenyl group of 11 resulted
in compound 12 (IC50 = 14 μM and 5.5 μM for ROCK1 and
ROCK2, respectively) with over an 11-fold improvement in
potency. Compound 12 was reported as a ROCK2 inhibitor
with an IC50 value of 260 nM along with three other analogues
of chloro- or fluoro-substituted phenyl groups.20,21 The 20−50-
fold difference in IC50 values between our compound 12 and
those reported20 is likely due to different assay conditions, that
is, the enzyme or ATP concentrations, and our emphasis is
relative activity. Because no compounds with a two carbon
spacer between the urea nitrogen and the phenyl group have
previously been reported, we then designed and synthesized six
novel compounds with a two carbon spacer (13−18).
Compound 18 showed an IC50 of 650 and 670 nM for
ROCK1 and ROCK2, respectively, with a 21- and 242-fold
improvement in potency (ROCK1) over compounds 12 and
11, respectively (Table 2), suggesting that the ethylene linker
distance allowed the phenyl group to reach a deep hydrophobic
pocket within the enzyme binding site of ROCK. However,
compound 19 with a three carbon spacer decreased the
inhibitory activities by 3.8-fold for ROCK1 and 1.6-fold for
ROCK2, respectively (IC50 = 2.46 and 1.07 μM for ROCK1
and ROCK2) with decreased LEs, suggesting that the ethylene
linker in 18 is the optimal spacer in this series. The addition of
methylenehydroxyl group to the ethylene as in 14 and hydroxyl
group to the ethylene in 15 or the addition of a methoxy or a
chloro to the phenyl ring as in 16 and 17 all diminished the
potency of 18 to varying degrees, further confirming the
hydrophobicity of the binding pocket. Replacement of the urea
hydrogen atom of 18 with a methyl group gave rise to the much
less potent compound 13, possibly due to the lack of essential
free hydrogen on one of the urea nitrogen atoms.
A molecular modeling study of a docked structure 18 into
ROCK1 (PDB code: 2ESM)22 suggested that potential
hydrogen-bonding interactions can be formed between amino
acid residues in the hinge region and 18. As shown in Figure 1a
in the Supporting Information, potential hydrogen-bonding
interactions are postulated between the indazole nitrogen and
the amide NH from M156, the indazole NH, and the carbonyl
O from E154 (this binding mode is consistent with that of a
cocrystal complex of a reported indazole-containing ROCK1
inhibitor23) and one of the urea NH and carboxylic O from
D216 of the DGF motif. However, R84, which is in close
proximity, was not utilized. To explore the possibility of this
additional hydrogen bonding, we considered a new fragment 4-
(pyridine-4-yl)aniline 20 as a surrogate of indazole 8. The
distance between the pyridyl nitrogen of 20 and the aniline
nitrogen connecting to the urea linker is longer (8.5 Å) than
that of 8 (5.9 Å). This shift may allow potential hydrogen-
bonding interactions with R84.
Six more compounds (22−27) with different spacers were
synthesized via the same route as that for synthesis of 10−19
described above, and their kinase inhibitory activities were
evaluated. By the time compounds 22−27 were synthesized,
this pyridine series was published in a patent, and most
recently, their anti-inflammatory activities as ROCK inhibitors
appeared.24 As shown in Table 3, compound 22 with a new
hinge binder moiety, 4-(pyridine-4-yl)aniline, showed similar
activity to that of 18 (Table 2). Compound 18 has a shorter
hinge binder (aminoindazole) but a longer spacer (ethylene)
while compound 22 has a longer hinge binder (pyridylaniline)
but a shorter spacer (methylene). Their in vitro activities
against ROCK1 and ROCK2 are from submicromolar to a
micromolar (ROCK1, 0.65 μM for 18 and 1.15 μM for 22;
ROCK2, 0.67 μM for 18 and 0.26 μM for 22; see Tables 2 and
3). As shown in Figure 1b in the Supporting Information,
compound 22 was docked in the same ATP binding pocket as
that of 18. While the hydrogen-bonding interaction with M156
is retained for 22, the interactions with E154 and D216 are lost.
New hydrogen-bonding interactions can be achieved between
the urea (NH and O) of 22 and R84.
Compounds 10−19 were synthesized from commercially
available 8 via a common intermediate 9, by reacting 8 with
phenylchloroformate in dichloromethane (DCM) to form an
activated carbamate followed by displacement by the
corresponding amines.
Most recently, we cocrystallized compound 18 with the
kinase domain of human ROCK1 (residues 6−415) to
2476
dx.doi.org/10.1021/jm201289r | J. Med. Chem. 2012, 55, 2474−2478

Journal of Medicinal Chemistry
Brief Article
experimentally determine the mode of action of our novel
indazole-containing ROCK1 inhibitor series. The ROCK1−18
complex crystallized in space group C2221 with two dimers per
asymmetric unit. The structure was refined to 2.3 Å resolution
with Rcryst and Rfree values of 18.8 and 23.6%, respectively
(Table 1 in the Supporting Information). The inhibitor binds
to the ATP site of ROCK1 essentially as predicted by molecular
docking (Figure 1c). The indazole nitrogen atoms establish
for ROCK1 over ROCK2. Eight-fold selectivity of ROCK1 over
ROCK2 is also observed for compound 25 with ethylene spacer
(Table 3).
The in vitro kinase SAR yielded potent and selective ROCK
inhibitors. We next determined whether some of these are
capable of entering intact cells, reaching their target and
inhibiting ROCK from phosphorylating its substrate MLC2. To
this end, we discovered that compound 18 and 24 inhibited
potently the phosphorylation of the ROCK substrate MLC2 in
intact human breast cancer cells as described in the Supporting
Information.
CONCLUSION
■
Recent studies identified ROCK inhibitors as potential
therapies for pathological conditions such as glaucoma.14−19
None of these studies have used FBDD approaches except for
the identification of a ROCK1 inhibitor from a historical
thrombin/FactorXa building block by fragment-based NMR
screening.25,26 In this study, using high concentration
biochemical assays and fragment-based screening, we have
discovered fragments to inhibit Rho-associated kinases. We also
demonstrated the design and optimization of ROCK inhibitors
using LE as a general guide to assess the binding potential of
the fragments and to guide the optimization process. Molecular
modeling aided the design and fragment hopping from one
hinge binder to another for the optimization of ROCK
inhibitors. Our structural biology studies yielded an X-ray
cocrystal of ROCK1−compound 18 in 2.3 Å resolution and,
coupled with molecular modeling studies, provided the
molecular basis for the design of more potent and selective
ROCK inhibitors. Optimization of fragments yielded potent
(100 nM) ROCK inhibitors that inhibited in intact human
cancer cells at low micromolar concentration the phosphor-
ylation of MLC2, a ROCK substrate, but not the phosphor-
ylation of proteins that are not substrates of ROCK such as
Erk1/2. Future studies will focus on determining the ability of
the most potent inhibitors to suppress migration and invasion,
cancer hallmarks known to be mediated by ROCK.
Figure 1. Molecular mode of action of compound 18. (a) Surface
presentation of the ROCK1 dimer in complex with 18 determined by
X-ray crystallography at 2.3 Å resolution. Exploded view detailing the
binding interactions of 18 (yellow) within the ATP site; the hinge and
DFG regions are indicated in cyan and orange, respectively. Displayed
in blue is the 2Fo − Fc electron density, contoured at 1σ around the
inhibitor. Potential hydrogen-bonding and van der Waal interactions
are shown as black and green dotted lines, respectively. (b) Schematic
presentation of the binding interactions between 18 and the ATP site.
(c) Overlay of compound 18 in the active site of ROCK1 determined
by X-ray crystallography (yellow) and predicted by molecular
modeling (green).
hydrogen-bonding interactions with the main chain of residues
Met156 and Glu154 of the hinge region. In addition, the
indazole ring establishes a series of van der Waals (hydro-
phobic) interactions with Ile82, Ala215, Ala103, and Leu205
(Figure 1b). The urea linker interacts with the side chain of
Asp216 of the DFG motif. The phenyl ring is sandwiched
between the P loop (Arg84−Gly88) of the upper N-terminal
lobe and the side chain of the catalytic residue Lys105 (Figure
1a). The crystal structure corroborates the results from the
molecular modeling studies, and only small changes are
observed in the conformation of the inhibitor molecule.
EXPERIMENTAL SECTION
■
The synthesis of fragments and ROCK inhibitors, molecular modeling,
X-ray cocrystallography, Z-lyte assays for determining ROCK kinase
activities, and effects of ROCK inhibitors on the phosphorylation
levels of MLC2 (a ROCK substrate) and Erk1/2 (not a ROCK
substrate) in human cancer cells are reported in the Supporting
Information.
ASSOCIATED CONTENT
■
S
* Supporting Information
Preparation of compounds 1−27, H NMR, HRMS, HPLC
Two pairs of chiral spacers were chosen to probe possible
stereochemical preferences that may exist in the enzyme
1
purity, Z-lyte assays, and effects of our ROCK inhibitors in
human cancer cells, molecular modeling, and X-ray cocrys-
tallography. This material is available free of charge via the
Internet at http://pubs.acs.org.
binding site. Compound 24 with an S-configuration (IC50
=
100 nM) showed 75-fold more kinase inhibitory activity than
compound 23 with a R-configuration (IC50 = 7520 nM) for
ROCK2, while the difference is only 5-fold for ROCK1 (IC50
=
9.07 μM for 23 vs 1.69 μM for 24). However, their
corresponding homologues with an additional methylene
spacer exhibited the opposite selectivity. While compound 26
with a R-configuration is only 6-fold more active than that of 27
with an S-configuration toward ROCK2 (IC50 = 5.36 μM for 26
vs 32.92 μM for 27), 22-fold more inhibitory activity toward
ROCK1 is observed (IC50 = 1.41 μM for 26 vs 31.01 μM for
27). In addition, compound 26 demonstrated 4-fold selectivity
AUTHOR INFORMATION
■
Corresponding Author
*Tel: 813-745-6485. Fax: 813-745-6875. E-mail: Rongshi.Li@
moffitt.org.
Present Address
§Department of Pharmacology, Penn State College of
Medicine, Hershey, Pennsylvania 17033.
2477
dx.doi.org/10.1021/jm201289r | J. Med. Chem. 2012, 55, 2474−2478

Journal of Medicinal Chemistry
Brief Article
S.; Mather, O.; Mercurio, C.; Minucci, S.; Montalbetti, C. A.; Muller,
A.; Patel, D.; Phillips, B. G.; Varasi, M.; Whittaker, M.; Winkler, D.;
Yarnold, C. J. Fragment-based identification of Hsp90 inhibitors.
ChemMedChem 2009, 4, 963−966.
(14) Dong, M.; Yan, B. P.; Liao, J. K.; Lam, Y. Y.; Yip, G. W.; Yu, C.
M. Rho-kinase inhibition: A novel therapeutic target for the treatment
of cardiovascular diseases. Drug Discovery Today 2010, 15, 622−629.
(15) LoGrasso, P. V.; Feng, Y. Rho kinase (ROCK) inhibitors and
their application to inflammatory disorders. Curr. Top. Med. Chem.
2009, 9, 704−723.
(16) Micuda, S.; Rosel, D.; Ryska, A.; Brabek, J. ROCK inhibitors as
emerging therapeutic candidates for sarcomas. Curr. Cancer Drug
Targets 2010, 10, 127−134.
(17) Yang, X.; Zhang, Y.; Wang, S.; Shi, W. Effect of fasudil on
growth, adhesion, invasion, and migration of 95D lung carcinoma cells
in vitro. Can. J. Physiol. Pharmacol. 2010, 88, 874−879.
(18) Kidera, Y.; Tsubaki, M.; Yamazoe, Y.; Shoji, K.; Nakamura, H.;
Ogaki, M.; Satou, T.; Itoh, T.; Isozaki, M.; Kaneko, J.; Tanimori, Y.;
Yanae, M.; Nishida, S. Reduction of lung metastasis, cell invasion, and
adhesion in mouse melanoma by statin-induced blockade of the Rho/
Rho-associated coiled-coil-containing protein kinase pathway. J. Exp.
Clin. Cancer Res. 2010, 29, 127.
ACKNOWLEDGMENTS
■
This work was supported partially by startup funds (R.L.),
5U19CA067771-15 (S.M.S). We thank the Moffitt Chemical
Biology Core facility for high concentration fragment-based
screening. We thank Drs. Bi-Cheng Wang and Lirong Chen at
the University of Georgia for the kind support during usage of
Beamline 22-ID, SER-CAT.
ABBREVIATIONS USED
■
FBDD, fragment-based drug discovery; HTS, high-throughput
screening; NMR, nuclear magnetic resonance; SPR, surface
plasmon resonance; ITC, isothermal titration calorimetry;
TINS, target immobilized NMR screening; NMS, native mass
spectrometry; WAC, weak affinity chromatography; ROCK,
Rho kinase; EDC, N-ethyl-N′-(3-dimethylaminopropyl)-
carbodiimide; DMAP, 4-N,N-dimethylaminopyridine; DMF,
N,N-dimethylformamide; LE, ligand efficiency; DCM, dichloro-
methane
(19) Hajduk, P. J. Fragment-based drug design: How big is too big? J.
Med. Chem. 2006, 49, 6972−6976.
(20) Iwakubo, M.; Takami, A.; Okada, Y.; Kawata, T.; Tagami, Y.;
Ohashi, H.; Sato, M.; Sugiyama, T.; Fukushima, K.; Iijima, H. Design
and synthesis of Rho kinase inhibitors (II). Bioorg. Med. Chem. 2007,
15, 350−364.
(21) Takami, A.; Iwakubo, M.; Okada, Y.; Kawata, T.; Odai, H.;
Takahashi, N.; Shindo, K.; Kimura, K.; Tagami, Y.; Miyake, M.;
Fukushima, K.; Inagaki, M.; Amano, M.; Kaibuchi, K.; Iijima, H.
Design and synthesis of Rho kinase inhibitors (I). Bioorg. Med. Chem.
2004, 12, 2115−2137.
(22) Jacobs, M.; Hayakawa, K.; Swenson, L.; Bellon, S.; Fleming, M.;
Taslimi, P.; Doran, J. The structure of dimeric ROCK I reveals the
mechanism for ligand selectivity. J. Biol. Chem. 2006, 281, 260−268.
(23) Sehon, C. A.; Wang, G. Z.; Viet, A. Q.; Goodman, K. B.;
Dowdell, S. E.; Elkins, P. A.; Semus, S. F.; Evans, C.; Jolivette, L. J.;
Kirkpatrick, R. B.; Dul, E.; Khandekar, S. S.; Yi, T.; Wright, L. L.;
Smith, G. K.; Behm, D. J.; Bentley, R.; Doe, C. P.; Hu, E.; Lee, D.
Potent, selective and orally bioavailable dihydropyrimidine inhibitors
of Rho kinase (ROCK1) as potential therapeutic agents for
cardiovascular diseases. J. Med. Chem. 2008, 51, 6631−6634.
(24) Yoshimi, E.; Kumakura, F.; Hatori, C.; Hamachi, E.; Iwashita, A.;
Ishii, N.; Terasawa, T.; Shimizu, Y.; Takeshita, N. Antinociceptive
effects of AS1892802, a novel Rho kinase inhibitor, in rat models of
inflammatory and noninflammatory arthritis. J. Pharmacol. Exp. Ther.
2010, 334, 955−963.
(25) Ray, P.; Wright, J.; Adam, J.; Bennett, J.; Boucharens, S.; Black,
D.; Cook, A.; Brown, A. R.; Epemolu, O.; Fletcher, D.; Haunso, A.;
Huggett, M.; Jones, P.; Laats, S.; Lyons, A.; Mestres, J.; de Man, J.;
Morphy, R.; Rankovic, Z.; Sherborne, B.; Sherry, L.; van Straten, N.;
Westwood, P.; Zaman, G. Z. Fragment-based discovery of 6-
substituted isoquinolin-1-amine based ROCK-I inhibitors. Bioorg.
Med. Chem. Lett. 2011, 21, 97−101.
REFERENCES
■
(1) Hajduk, P. J.; Greer, J. A decade of fragment-based drug design:
strategic advances and lessons learned. Nat. Rev. Drug Discovery 2007,
6, 211−219.
(2) Bohacek, R. S.; McMartin, C.; Guida, W. C. The art and practice
of structure-based drug design: a molecular modeling perspective. Med.
Res. Rev. 1996, 16, 3−50.
(3) Fink, T.; Reymond, J. L. Virtual exploration of the chemical
universe up to 11 atoms of C, N, O, F: assembly of 26.4 million
structures (110.9 million stereoisomers) and analysis for new ring
systems, stereochemistry, physicochemical properties, compound
classes, and drug discovery. J. Chem. Inf. Model. 2007, 47, 342−353.
(4) Jhoti, H.; Cleasby, A.; Verdonk, M.; Williams, G. Fragment-based
screening using X-ray crystallography and NMR spectroscopy. Curr.
Opin. Chem. Biol. 2007, 11, 485−493.
(5) Navratilova, I.; Hopkins, A. L. Fragment Screening by Surface
Plasmon Resonance. ACS Med. Chem. Lett. 2010, 1, 44−48.
(6) Ladbury, J. E.; Klebe, G.; Freire, E. Adding calorimetric data to
decision making in lead discovery: a hot tip. Nat. Rev. Drug Discovery
2010, 9, 23−27.
(7) Siegal, G.; Hollander, J. G. Target immobilization and NMR
screening of fragments in early drug discovery. Curr. Top. Med. Chem.
2009, 9, 1736−1745.
(8) Hannah, V. V.; Atmanene, C.; Zeyer, D.; Dorsselaer, A. V.;
́
Sanglier-Cianferani, S. Native MS: An 'ESI, way to support structure-
and fragment-based drug discovery. Future Med. Chem. 2010, 2, 35−
50.
(9) Duong-Thi, M. D.; Meiby, E.; Bergstrom, M.; Fex, T.; Isaksson,
R.; Ohlson, S. Weak affinity chromatography as a new approach for
fragment screening in drug discovery. Anal. Biochem. 2011, 414, 138−
146.
(10) Medina, J. R.; Blackledge, C. W.; Heerding, D. A.; Campobasso,
N.; Ward, P.; Briand, J.; Wright, L.; Axten, J. M. Aminoindazole PDK1
Inhibitors: A Case Study in Fragment-Based Drug Discovery. ACS
Med. Chem. Lett. 2010, 1, 439−442.
(11) Godemann, R.; Madden, J.; Kramer, J.; Smith, M.; Fritz, U.;
Hesterkamp, T.; Barker, J.; Hoppner, S.; Hallett, D.; Cesura, A.;
Ebneth, A.; Kemp, J. Fragment-based discovery of BACE1 inhibitors
using functional assays. Biochemistry 2009, 48, 10743−10751.
(12) Matthews, T. P.; Klair, S.; Burns, S.; Boxall, K.; Cherry, M.;
Fisher, M.; Westwood, I. M.; Walton, M. I.; McHardy, T.; Cheung, K.
M.; Van Montfort, R.; Williams, D.; Aherne, G. W.; Garrett, M. D.;
Reader, J.; Collins, I. Identification of inhibitors of checkpoint kinase 1
through template screening. J. Med. Chem. 2009, 52, 4810−4819.
(13) Barker, J. J.; Barker, O.; Boggio, R.; Chauhan, V.; Cheng, R. K.;
Corden, V.; Courtney, S. M.; Edwards, N.; Falque, V. M.; Fusar, F.;
Gardiner, M.; Hamelin, E. M.; Hesterkamp, T.; Ichihara, O.; Jones, R.
(26) Ray, P.; Wright, J.; Adam, J.; Boucharens, S.; Black, D.; Brown,
A. R.; Epemolu, O.; Fletcher, D.; Huggett, M.; Jones, P.; Laats, S.;
Lyons, A.; de Man, J.; Morphy, R.; Sherborne, B.; Sherry, L.; Straten,
N.; Westwood, P.; York, M. Optimisation of 6-substituted isoquinolin-
1-amine based ROCK-I inhibitors. Bioorg. Med. Chem. Lett. 2011, 21,
1084−1088.
2478
dx.doi.org/10.1021/jm201289r | J. Med. Chem. 2012, 55, 2474−2478