Triazine and pyrimidine based ROCK inhibitors with efficacy in spontaneous hypertensive rat model

Article abstract of DOI:10.1016/j.bmcl.2009.09.046

The profile of a series of triazine and pyrimidine based ROCK inhibitors is described. An initial binding mode was established based on a homology model and the proposed interactions are consistent with the observed SAR. Compounds from the series are pote

Full text of DOI:10.1016/j.bmcl.2009.09.046

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6027–6031
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Triazine and pyrimidine based ROCK inhibitors with efficacy
in spontaneous hypertensive rat model
a
a
b
a
a
Koc-Kan Ho a, , James R. Beasley , Laura Belanger , Darcey Black , Jui-Hsiang Chan , David Dunn ,
Bing Hu c, Anthony Klon a, Steven G. Kultgen a, Michael Ohlmeyer a, Susan M. Parlato a, Peter C. Ray b,
Quynhchi Pham a, Yajing Rong a, Andrew L. Roughton a, Tiffany L. Walker a, Jane Wright b, Kai Xu c, Yan Xu c,
Limei Zhang c, Maria Webb a
*
a Ligand Pharmaceuticals, Inc., 3000 Eastpark Boulevard, Cranbury, NJ 08512, USA
b Schering-Plough Research Institute, Newhouse, Lanarkshire, ML1 5SH, UK
c WuXi AppTec Co., Ltd, 288 Fute Zhong Road, Shanghai 200131, China
a r t i c l e i n f o
a b s t r a c t
Article history:
The profile of a series of triazine and pyrimidine based ROCK inhibitors is described. An initial binding
mode was established based on a homology model and the proposed interactions are consistent with
the observed SAR. Compounds from the series are potent in a cell migration assay and possess a favorable
kinase selectivity. In vivo activity was demonstrated for compound 1A in a spontaneous hypertensive rat
model.
Received 18 August 2009
Revised 11 September 2009
Accepted 14 September 2009
Available online 17 September 2009
Ó 2009 Published by Elsevier Ltd.
Keywords:
ROCK
Triazine
Pyrimidine
Homology model
Hypertension
Rho-kinase (ROCKI and ROCKII), a downstream effector of RhoA
G protein, is one of the key mediators in regulating vascular
smooth muscle tone.1,2 Activation of ROCK by pathological agents,
such as angiotensin II, endothelin-1, 5-hydroxytryptamine and
oxyhemoglobin, leads to inhibition of myosin light chain phospha-
tase, and in turn increases the vascular tone.1 Fasudil, a ROCK
inhibitor and a dilator for injured cerebral arteries, has been used
clinically in Japan since 1995 for the treatment of cerebral vaso-
spasm.3 Since then, pharmaceutical companies have been investi-
gating small molecule ROCK inhibitors as potential treatments of
hypertension and related cardiovascular disease.4
ROCK also plays an important role in cellular functions such as
cell motility and cell migration.5 Inhibition of ROCK may stabilize le-
sions by reducing monocyte infiltration directly by inhibiting the
motility of monocytes, and indirectly by preventing the increase in
endothelial cell permeability. Therefore, the modulation of Rho-ki-
nase activity could be a useful treatment for atherosclerosis.6
A high-throughput screening campaign was initiated to iden-
tify ROCK inhibitors. An IMAPÒ,7 based ROCKI assay was used to
screen Pharmacopeia libraries (5 million compounds, 2005). One
of the hit series represented by structure 1A was identified and
was subsequently demonstrated to possess excellent cell based
activity in a Monocyte Chemotactic Protein-1 (MCP-1) induced
THP-1 cell migration assay (Table 1). Since compound 1A con-
tains a triaminotriazine motif which is common to many kinase
inhibitors, it was a priority to assess its general kinase selectiv-
ity. Thus, compound 1A was tested against a panel of 24 kinase
at a concentration of 2 lM. Compound 1A shows an acceptable
selectivity profile (>25-fold) against the kinase panel.8
This Letter describes the structural–activity relationships around
the compound and a general binding mode is proposed for the series
based on a homology model.
Triazine and pyrimidine analogs were readily accessible via
the synthetic route outlined in Schemes 1 and 2. It is based on
sequential addition of amines to cyanuric chloride or trichloro-
pyrimidine. One key step involves the displacement of the chlo-
ride in a diaminochlorotriazine (1D to 1E) with aminopyridines.
Although
a direct displacement with 4-aminopyridine under
strong heating condition provided sufficient quantity for the ini-
tial analoging, similar conditions used to attach 3-aminopyridine
and 2-aminopyridine to the chlorotriazine were met with low
yield and impure products. Subsequently, a palladium mediated
* Corresponding author. Tel.: +1 609 452 3695; fax: +1 609 452 3671.
E-mail address: kho@ligand.com (K.-K. Ho).
In 2008, Pharmacopeia, Inc was acquired by Ligand Pharmaceuticals, Inc.
0960-894X/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2009.09.046

6028
K.-K. Ho et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6027–6031
Table 1
zyme interaction. The exocyclic NH in this domain does not appear
to be involved in any binding interaction since an ether analog (3F)
is equal potent to the N-linked 1A.
Profile of hit 1A identified in a ROCKI screening campaign
The 3-pyridyl (4A) and 2-pyridyl (4B) isomers were prepared in
order to establish the optimal position of the pyridyl nitrogen in
the R2 domain (Table 3). Significant decrease in potency in both
cases suggested that the endocyclic pyridyl nitrogen may involve
in a hydrogen bonding interaction. The hypothesis is further
strengthened by the fact that an aniline analog (4C) has inferior
potency.
NH
N
N
N
NH
N
N
NH
1A
a
Next, the piperidine domain (R3, Table 4) was evaluated. The ba-
sic nitrogen distal from the triazine is required for potency. Dele-
tion (5A) or methylation (5B) of the distal nitrogen yielded
weaker compounds (R3, Table 4). Cyclic amines (1A and 5C) are
preferred over the linear aliphatic amine (5D). Four methylpipri-
dine analogs were prepared (5E to 5H) in order to elucidate the
preferred position and chirality for the methyl substitution. Among
the four isomers, compound 5E shows superior potency in both the
IMAP assay and cell migration assays.
Assays/parameters
IC50 (nM)
ROCKI (IMAP)
THP-1 (MCP-1)
Selectivity (24 kinases)
MW
6 ( 2)
30 ( 1)
<50% inhibition at 2 lM
405
A log P
3.71
a
Values are means of two or more measurements, standard deviation is given in
parentheses (na = not active).
We next turned our attention to the triazine core. The core in 1A
was replaced by the three possible pyrimidine isomers (Table 5).
Stepwise replacement of a nitrogen in the core with a CH estab-
lished that the nitrogen between the 4-aminopyridine and piperi-
zine moieties (position Z in general structure 6) is required for
potency. Thus, compound 6C, in which the Z group is a CH, shows
significantly weaker potency in comparison to 6A and 6B. We spec-
ulated that the nitrogen may involve in a specific interaction with
the enzyme.
coupling procedure was used to prepare the 3-aminopyridyl (4A)
and 2-aminopyridyl (4B) analogs.
The SAR was first established for the 2-aminoindane moiety (R1,
Table 2). Weaker potency was observed for the cyclopentylamino
analog (3A) and the data suggested that an aromatic group is re-
quired for potency. The orientation of the aromatic moiety for opti-
mal interaction was explored by two ring opening analogs (3B and
3C), a 2-aminotetrahydronaphthalene analog (3D) and a 1-amino-
indane analog (3E). The superior potency observed in 1A in
comparison to these compounds showed that the fused ring in
the 2-aminoindane imposed a favorable geometry for ligand en-
In the absence of a ROCK crystal structure (year 2005),9 a homol-
ogy model of ROCKI was built with MOE (Chemical Computing
Cl
Cl
HN
NBoc
NH2
NH
N
N
N
N
N
N
N
Cl
N
N
NBoc
N
NBoc
a
b
N
Cl
Cl
Cl
1B
1C
1D
NH
N
NH
N
N
NH2
N
N
N
N
N
NBoc
N
NH HCl
c
d
N
N
NH
N
NH
1E
1A
O
O
N
N
NH2
OH
N
N
N
NH TFA
N
NBoc
N
1C
1D
N
f, g
N
NH
e
Cl
2F
1F
NH2
NH
N
NH
N
for 3A
for 3B
N
N
N
N
N
N
NBoc
N
NH TFA
i
NH2
h
R2
R2
N
1G, R2 = 3-aminopyridyl
1H, R2 = 2-aminopyridyl
4A, R2 = 3-aminopyridyl
4B, R2 = 2-aminopyridyl
Scheme 1. Synthesis of triazine based ROCK inhibitors. Reagents and conditions: (a) NaHCO3, acetone, H2O, 0–5 °C; (b) CH2Cl2, NEt3, rt; (c) K2CO3, THF, 140 °C, sealed-tube;
(d) HCl in dioxane; (e) NaOH, acetone, H2O; (f) K2CO3, DMSO, 80 °C; (g) TFA, CH2Cl2; (h) Pd(OAc)2, BINAP, Cs2CO3, dioxane, microwave target temperature 160 °C; (i) TFA,
CH2Cl2.

K.-K. Ho et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6027–6031
6029
Cl
Cl
Cl
Cl
Cl
N
HN
NBoc
N
N
N
N
N
Cl
+
N
NBoc
BocN
N
a
Cl
2A
2B
2C
NH
NH
N
NH2
NH
NH2
N
N
N
2B
N
NH TFA
N
NBoc
c, d
b
N
N
Cl 2D
6A
Cl
NH
N
N
NH2
NH2
N
N
NBoc
N
NH TFA
2C
2A
N
e
N
f, g
N
NH
N
NH
2E
6B
NH2
NH
NH
N
N
N
+
N
Cl
Cl
h
Cl
Cl
2F
2G
NH
N
HN
NBoc
N
NH2
NH
N
N
2F
N
NH TFA
N
N
NBoc
j, k
i
NH6C
N
Cl
Scheme 2. Synthesis of pyrimidine based ROCK inhibitors. Reagents and conditions: (a) DIEA, CH2Cl2, À78 °C, then separate by column chromatography; (b) K2CO3, CH3CN,
reflux; (c) Pd(OAc)2, BINAP, Cs2CO3, dioxane, microwave target temperature 160 °C; (d) TFA, CH2Cl2; (e) K2CO3, CH3CN, reflux; (f) K2CO3, THF, reflux; (g) TFA, CH2Cl2; (h)
CH2Cl2, À78 °C, then separate by column chromatography; (i) K2CO3, CH3CN, reflux; (j) Pd(OAc)2, BINAP, Cs2CO3, dioxane, microwave target temperature 160 °C; (k) TFA,
CH2Cl2.
Group, Montreal, Quebec, Canada, version 2006.08) using a crystal
structure of PKA (1Q8T) as a template.10 Docking of 1A was per-
formed with Glide in standard precision (SP) mode.11 Key interac-
tions between the ligand and the enzyme are shown in Figure 1
and are consistent with the observed SAR.
In the proposed binding model for 1, the 2-aminoindane moiety
interacts with ROCKI via hydrophobic contact with the aromatic
ring of Phe-368. The nitrogen has no observed role in forming spe-
cific hydrogen bonding interactions, a conclusion that is supported
by the observed equipotency of compound 3F with 1A (Table 2).
Akeyintermolecularhydrogenbondisobservedbetweenthepyr-
idyl nitrogen and the backbone NH of Met-156 in the hinge domain.
The model suggests that the geometric orientation of this contact is
critical for binding to ROCKI, and this is borne out in the SAR by sub-
stantial losses in potency when the location of the nitrogen on the
ring is altered (4A and 4B) or is removed altogether (4C).
The triazine nitrogen between the 4-aminopyridine and piperi-
zine moieties (6C) does not appear to have a specific role in the
proposed binding mode. The loss of potency of replacing this nitro-
gen can be rationalized in one of two ways. Either there is enough
conformational flexibility in the Lys-105 side chain to accommo-
date a direct interaction between the side chain nitrogen with
the core nitrogen of 6C (distance 5.6 Å), or a water-mediated
hydrogen bond is necessary for interactions with hydrogen bond
acceptors/donors as represented by Lys-105. The other two nitro-
gens in the triazine core have no role in the proposed binding mod-
el, which is consistent with the observation that their removal has
little effect on their IMAPÒ potency (6A and 6B).
A representative example from the series (1A) was tested in a
spontaneous hypertensive rat model. Compound was dosed via
sc in three doses (1, 3 and 10 mg/kg).14 A significant decrease in
mean arterial blood pressure (MAP) was observed at 3 and
10 mg/kg with an apparent dose–response (Fig. 2).
The model for the binding mode of 1A places the basic nitrogen
of the piperidine ring in position to form a hydrogen bond with one
of the side chain oxygen atoms of Asp-216. Removal of this hydro-
gen bonding interaction yielded substantially weaker compounds
(5A and 5B).
In conclusion, triazine and pyrimidine based ROCK inhibitors
were identified. An initial binding mode was established based
on a homology model and the proposed interactions are consis-
tence with the observed SAR. Compounds from the series are

6030
K.-K. Ho et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6027–6031
Table 4
Table 2
Modifications of R3
Modifications of R1
R1
N
NH
N
N
N
N
N
N
NH
R3
N
NH
N
NH
3
5
Compds R1
ROCKI (IMAP) IC50
(nM)
THP-1 (MCP-1) IC50
(nM)
a
a
Compds R3
ROCKI (IMAP) IC50
(nM)
THP-1 (MCP-1) IC50
(nM)
a
a
1A
3A
3B
3C
3D
3E
6 ( 2)
30 ( 1)
nd
NH
1A
6 ( 2)
30 ( 1)
N
N
N
N
NH
H
N
890 ( 315)
326 ( 3)
400 ( 6)
23 ( 1)
5A
5B
5C
5D
5E
5F
5G
5H
8300 ( 4900)
305 ( 14)
11 ( 1)
nd
H
N
nd
nd
N
H
nd
N
H
nd
N
17 ( 10)
NH
NH
NH
H
N
185 ( 93)
8 ( 2)
nd
81 ( 2)
nd
N
N
3F
57 ( 30)
O
R
S
8 ( 2)
26 ( 13)
a
Values are means of two or more measurements, standard deviation is given in
parentheses (nd = not determined).
N
N
N
NH
NH
NH
135 ( 57)
118 ( 6)
87 ( 7)
nd
Table 3
Modifications of R2
nd
R
S
NH
N
N
N
NH
N
nd
R2
4
a
Values are means of two or more measurements, standard deviation is given in
parentheses (nd = not determined).
Compds R2
ROCKI (IMAP) IC50
(nM)
THP-1 (MCP-1) IC50
(nM)
a
a
NH
NH
NH
N
1
6 ( 2)
30 ( 1)
Table 5
Replacement of triazine with pyrimidines
4A
4B
4C
8400 ( 200)
1000 ( 100)
7400 ( 900)
nd
nd
nd
N
NH
Y
Z
X
N
NH
N
N
NH
6
NH
a
a
Compds
X
Y
Z
ROCKI (IMAP) IC50
(nM)
THP-1 (MCP-1) IC50
(nM)
a
Values are means of two or more measurements, standard deviation is given in
parentheses (nd = not determined).
1A
6A
6B
6C
N
CH
N
N
N
CH
N
N
N
N
6 ( 2)
2 ( 1)
14 ( 4)
30 ( 1)
12 ( 2)
154 ( 53)
nd
N
CH 1200 ( 120)
potent in a cell migration assay and possess a favorable kinase
selective. In vivo activity was demonstrated in the spontaneous
hypertensive rat model.
a
Values are means of two or more measurements, standard deviation is given in
parentheses (nd = not determined).

K.-K. Ho et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6027–6031
6031
References and notes
1. Budzyn, K.; Marley, P. D.; Sobey, C. G. Trends Pharmacol. Sci. 2006, 27, 97.
2. Lee, D. L.; Webb, R. C.; Jin, L. Hypertension 2004, 44, 796.
3. Mueller, B. K.; Mack, H.; Teusch, N. Nat. Rev. Drug Disc. 2005, 4, 387.
4. Hu, E.; Lee, D. Exp. Opin. Ther. Targets 2005, 9, 4; Stavenger, R. A.; Cui, H.;
Dowdell, S. E.; Franz, R. G.; Gaitanopoulos, D. E.; Goodman, K. B.; Hilfiker, M. A.;
Ivy, R. L.; Leber, J. D.; Marino, J. P., Jr.; Oh, H.-J.; Viet, A. Q.; Xu, W.; Ye, G.; Zhang,
D.; Zhao, Y.; Jolivette, L. J.; Head, M. S.; Semus, S. F.; Elkins, P. A.; Kirkpatrick, R.
B.; Dul, E.; Khandekar, S. S.; Yi, T.; Jung, D. K.; Wright, L. L.; Smith, G. K.; Behm,
D. J.; Doe, C. P.; Bentley, R.; Chen, Z. X.; Hu, E.; Lee, D. J. Med. Chem. 2007, 50, 2;
Goodman, K. B.; Cui, H.; Dowdell, S. E.; Gaitanopoulos, D. E.; Ivy, R. L.; Sehon, C.
A.; Stavenger, R. A.; Wang, G. Z.; Viet, A. Q.; Xu, W.; Ye, G.; Semus, S. F.; Evans,
C.; Fries, H. E.; Jolivette, L. J.; Kirkpatrick, R. B.; Dul, E.; Khandekar, S. S.; Yi, T.;
Jung, D. K.; Wright, L. L.; Smith, G. K.; Behm, D. J.; Bentley, R.; Doe, C. P.; Hu, E.;
Lee, D. J. Med. Chem. 2007, 50, 6; Schirok, H.; Kast, R.; Figueroa-Pérez, S.;
Bennabi, S.; Gnoth, M. J.; Feurer, A.; Heckroth, H.; Thutewohl, M.; Paulsen, H.;
Knorr, A.; Hütter, J.; Lobell, M.; Münter, K.; Geiß, V.; Ehmke, H.; Lang, D.;
Radtke, M.; Mittendorf, J.; Stasch, J.-P. ChemMedChem 2008, 12, 1893.
5. Saito, H.; Minamiya, Y.; Saito, S.; Ogawa, J. J. Leukocyte Biol. 2002, 72, 829.
6. Shimokawa, H.; Morishige, K.; Miyata, K.; Kandabashi, T.; Eto, Y.; Ikegaki, I.;
Asano, T.; Kaibuchi, K.; Takeshita, A. Cardiovasc. Res. 2001, 51, 169; Morishige,
K.; Shimokawa, H.; Eto, Y.; Kandabashi, T.; Miyata, K.; Matsumoto, Y.;
Hoshijima, M.; Kaibuchi, K.; Takeshita, A. Arterioscler. Thromb. Vasc. Biol.
2001, 21, 548.
7. Registered trademark of Molecular Devices Sportsman, J. R.; Gaudet, E. A.; Boge,
A. Assay Drug Dev. Technol. 2004, 2, 205.
8. The following kinases were screened by Upstate (Millipore, MA, USA) at 2
lM
concentration: EGFR, IR, LKB1, MEK1, PDGFRb, TrkB, AuroraA, CaMKIV, CDK2,
CHK1, CK1d, CK2, cRAF, CSK, cSRC, GSK3b, LCK, MAPK2, MAPKA, Met, NEK2,
PAK2, Rsk2 and Zap-70.
9. The modeling study was conducted in 2005. A ROCK crystal structure was
published in 2006 Jacobs, M.; Hayakawa, K.; Swenson, L.; Bellon, S.; Fleming,
M.; Taslimi, P.; Doran, J. J. Biol. Chem. 2006, 281, 260.
10. Breitenlechner, C.; Gassel, M.; Hidaka, H.; Kinzel, V.; Huber, R.; Engh, R. A.;
Bossemeyer. Structure 2003, 11, 1595.
11. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.;
Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.;
Shenkin, P. S. J. Med. Chem. 2004, 47, 1739; Halgren, T. A.; Murphy, R. B.;
Friesner, R. A.; Beard, H. S.; Frye, L. L.; Pollard, W. T.; Banks, J. L. J. Med. Chem.
2004, 47, 1750.
12. Clark, A. M.; Labute, P.; Santavy, M. J. Chem. Inf. Model. 2006, 46, 1107.
13. DeLano, W. L. DeLano Scientific: Palo Alto, CA, USA, 2002, http://
www.pymol.org.
14. Y-27632,
a known ROCK inhibitor, was used as positive control in the
spontaneous hypertensive rat model. Y-27632 was dosed orally at 30 mg/kg
and significant decrease in MAP was observed. The study was conducted at
Aptuit (CT, USA).
Figure 1. Interaction of 1 with the ROCK1 homology model. The top panel was
generated with MOE.12 The bottom panel was generated with PyMOL.13 The distal
nitrogen of the piperzine is in quaternary form. One of the hydrogen was deleted for
clarity reason.
Figure 2. Effect of compound 1A in mean arterial blood pressure in a spontaneous
hypertensive rat model (1, 3 and 10 mg/kg sc).