Substituted 2H-isoquinolin-1-ones as potent Rho-kinase inhibitors: Part 3, aryl substituted pyrrolidines

Article abstract of DOI:10.1016/j.bmcl.2010.04.069

The discovery and SAR of a series of β-aryl substituted pyrrolidine 2H-isoquinolin-1-one inhibitors of Rho-kinase (ROCK) derived from 2 is herein described. SAR studies have shown that aryl groups in the β-position are optimal for potency. Our efforts foc

Full text of DOI:10.1016/j.bmcl.2010.04.069

Bioorganic & Medicinal Chemistry Letters 20 (2010) 3746–3749
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Substituted 2H-isoquinolin-1-ones as potent Rho-kinase inhibitors: Part 3,
aryl substituted pyrrolidines
a
a
a
b
a
Todd Bosanac a, , Eugene R. Hickey , John Ginn , Mohammed Kashem , Steven Kerr , Stanley Kugler ,
*
Xiang Li a, Alan Olague a, Sabine Schlyer a, Erick R. R. Young a
a Department of Medicinal Chemistry, Boehringer-Ingelheim Pharmaceuticals, Inc., 900 Ridgebury Road, PO Box 368, Ridgefield, CT 06877-0368, United States
b Department of Cardiometabolic Diseases, Boehringer-Ingelheim Pharmaceuticals, Inc., 900 Ridgebury Road, PO Box 368, Ridgefield, CT 06877-0368, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The discovery and SAR of a series of b-aryl substituted pyrrolidine 2H-isoquinolin-1-one inhibitors of
Rho-kinase (ROCK) derived from 2 is herein described. SAR studies have shown that aryl groups in the
b-position are optimal for potency. Our efforts focused on improving the ROCK potency of this isoquino-
lone class of inhibitors which led to the identification of pyrrolidine 32 which demonstrated a 10-fold
improvement in aortic ring (AR) potency over 2.
Received 24 February 2010
Revised 13 April 2010
Accepted 16 April 2010
Available online 21 April 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keyword:
Rho-kinase
The serine–threonine Rho-kinases, ROCK1 and ROCK2, have
been repeatedly reviewed as desirable points of therapeutic
intervention for many disease indications.1 With respect to cardio-
vascular disease they are known to prolong smooth muscle con-
traction through an inactivating phosphorylation of myosin light
chain kinase,2 inhibit endothelium mediated smooth muscle relax-
ation through reducing bioavailability of NO,3 play an essential role
in proinflammatory chemotactic cell migration,4 and in some con-
texts may be profibrotic.5 For these reasons, we became interested
in the inhibition of Rho-kinase for treatment of hypertension with
independent add on benefits for end organ protection.
N
O
O
N
H
NH
Cl
NH
Cl
HN
HN
O
O
1
2
ROCK2 IC50 = 12 nM
AR EC50 = 170 nM
HLM = 15 min
ROCK2 IC50 = 31 nM
AR EC50 = 250 nM
HLM = >300 min
Recently we described the discovery of a new class of 2H-iso-
quinolin-1-one based ROCK1/ROCK2 dual inhibitors6 along with
our early optimization efforts to identify analogs demonstrating
sustained blood pressure lowering in the spontaneous hyperten-
sive rat model.7 A key finding for achieving oral exposure with this
series was the flexibility to sacrifice the metabolically labile phenyl
glycine side chain of 1 (Fig. 1). This was realized upon further opti-
mizing potency through the basic amine-phosphate binding region
interaction leading to the identification of 2.
We had previously observed that the aryl group of the phenyl
glycine residue contributes up to a 10-fold improvement in binding
potency when compared to an unsubstituted analog (Table 1).
Docking experiments with a ROCK1 homology model reveal the
likely source of this binding interaction is extension of the aryl sub-
Figure 1.
the glycine rich loop vide infra. The goal of the work described
herein was to reincorporate this binding interaction into piperidine
analog 2 without compromising metabolic stability and oral
bioavailability.
Based upon binding overlays of 2 and 4 we reasoned that
either the
a or b ring carbon of piperidines or related pyrrolidines
might provide the appropriate trajectory to access the G-loop
hydrophobic pocket without disrupting the equally important
amine interaction with residues in the phosphate binding region
(Fig. 2).
Substituted analogs of 2 at the
from their corresponding N-Boc protected acids 5 (Scheme 1).8 The
-analogs were prepared via conversion to the acid chloride fol-
a- and b-carbons were prepared
stituent from the
a-carbon into a hydrophobic groove defined by
a
lowed by coupling with isoquinolone 6 and removal of the protect-
* Corresponding author.
ing group. Alternatively, the b-analogs were synthesized using the
E-mail address: todd.bosanac@boehringer-ingelheim.com (T. Bosanac).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.04.069

T. Bosanac et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3746–3749
3747
Table 1
Table 2
Activity of 2H-isoquinolin-1-ones
SAR of substituted piperidines and pyrrolidines
O
O
R
R
NH
NH
HN
HN
Cl
Cl
O
O
a
ROCK2a IC50 (nM) AR EC50 (nM) HLMc (min)
b
Compd
R
ROCK2 IC50 (nM)
Compd
R
NH
NH
3
120
NH2
9
28
67
47
220
nt
224
>300
71
Cl
10
11
4
15
NH2
550
NH
NH
a
Cambrex PKLight ATP detection reagent using luciferin–luciferase to quantify
residual ATP. Values are means of at least two duplicate experiments.
12
13
230
240
nt
nt
nt
nt
NH
H
N
14-rac
6.0
55
37
15-rac
6.0
89
130
NH
a
Cambrex PKLight ATP detection reagent using luciferin–luciferase to quantify
residual ATP. Values are means of at least two duplicate experiments.
b
Relaxation of phenylephrine stimulated isolated rat aortic rings. Values are
means of at least three experiments.
c
Compounds were incubated with human liver microsomes at a concentration of
Figure 2. Overlay of inhibitor 2 (salmon) and 4 (blue) in a ROCK homology model.
1 mg of protein/mL. nt = not tested.
Table 3
SAR of substituted phenyl pyrrolidines
R
NBoc
R
NH
O
O
n
O
NH
NH
Cl
1) SOCl2, TEA
R
N
O
HN
HN
O
Cl
HCl
2) 6, TEA, DMF
n
O
O
OH
7
5
6
Compd
R
ROCK2 IC50
(nM)
AR EC50
(nM)
HLMc
(min)
a
b
R
NH
Br
O
O
n
NH2
Cl
16-rac
2.0
63
43
NH
Cl
1) 5, POCl3, Pyr
HN
HN
2) HCl
O
O
8
17-rac
18-rac
19-rac
4.0
5.0
4.0
88
nt
nt
30
nt
nt
Scheme 1. Synthesis of isoquinolone inhibitors.
phosphorous oxychloride/pyridine conditions described previously
to effect coupling.7
Our initial array of targets focused on identifying the optimal
heterocycle and substitution pattern. Compounds were first pro-
filed for their inhibition of ROCK 2 using a luciferase based ATP
detection assay.6 Compounds that displayed sufficient activity
(continued on next page)

3748
T. Bosanac et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3746–3749
improved aortic ring potency. Pyrrolidines were prioritized for
Table 3 (continued)
further exploration with the discovery that 15 successfully main-
tained good microsomal stability and displayed higher oral bio-
availability in rat (97% vs 13% compared to piperidine analog 14).10
Substitution around the phenyl ring of 15 was further explored
and both electron donating and withdrawing groups were found to
be well tolerated with respect to molecular potency (Table 3). Ini-
tial exploration of substitution focused on the 3-position which
demonstrated good aortic ring potency (3-Br and 3-OMe substi-
tuted analogs 16 and 17), however poor microsomal stability was
observed with these inhibitors. We reasoned that metabolism
was occurring on the aryl ring and chose to investigate both
replacement of the phenyl group with a pyridine, an effective strat-
egy to alter metabolism,11 and halogenation at the 4-position. The
pyridyl compounds 25–27 did not improve the microsomal stabil-
ity. Significant differences in the microsomal stability of the 4-hal-
ogenated derivatives were observed (compare 28–30). In particular
the 4-Cl compound 30 demonstrated both the best aortic ring po-
tency and HLM stability and was resolved through chiral hplc to af-
ford the discrete enantiomers 31 and 32.12 The absolute
stereochemistry of 31 was determined by an x-ray co-crystal struc-
ture with ROCK1 (vide infra).
a
b
Compd
R
ROCK2 IC50
(nM)
AR EC50
(nM)
HLMc
(min)
20-rac
6.0
11
nt
nt
nt
nt
21-rac
22-rac
OCF3
CF3
7.0
nt
nt
23-rac
24-rac
4.0
8.0
nt
nt
nt
nt
Cl
Cl
A modest separation in potency was seen with the (3S,4R) iso-
mer 32 being more potent. The activity of both 31 and 32 could
be rationalized with our ROCK1 homology model (Fig. 3) in which
both compounds are capable of simultaneously participating in
both the targeted G-loop and phosphate binding region
interactions.
N
25-rac
26-rac
27-rac
14
8.0
26
210
150
66
58
49
nt
N
A co-crystal structure of inhibitor 31 bound to ROCK1 was
solved at 3.2Å resolution (Fig. 4), which revealed a hydrogen bond
between the pyrrolidine nitrogen and the carboxylate of ASP202,
consistent with the docked structure of 31 in the ROCK homology
model. The co-crystal structure also shows the 4-Cl-phenyl group
is inserted right beneath the G-loop making hydrophobic contacts
with Leu107 and Lys105. The G-loop is significantly shifted (about
1.8 Å) compared to the co-structure with inhibitor 22 described
previously7 in order to accommodate the phenyl group of 31.
In conclusion, we have designed a series of b-aryl-pyrrolidine
2H-isoquinolin-1-one dual ROCK2 inhibitors which we believe ac-
cess both the G-loop lipophilic interaction of the lead phenyl gly-
cine series and the phosphate binding region polar interaction of
the isonipocotic acid derivatives.7 Execution of this strategy
accomplished the goal of improving aortic ring potency over the
previously reported 2, 10-fold while maintaining excellent micro-
somal stability.
N
F
28-rac
29-rac
9.0
4.0
6.0
160
140
60
61
Br
Cl
238
>300
30-rac
Cl
Cl
31-
(3R,4S)
11
40
23
nt
nt
32-
(3S,4R)
4.0
a
Cambrex PKLight ATP Detection Reagent using luciferin–luciferase to quantify
residual ATP. Values are means of at least two duplicate experiments.
b
Relaxation of phenylephrine stimulated isolated rat aortic rings. Values are
means of at least three experiments.
c
Compounds were incubated with human liver microsomes at a concentration of
1 mg of protein/mL. nt = not tested.
against ROCK 2 were further examined for dilation of isolated rat
aortic rings following pre-constriction with phenylephrine.9 The
initial heterocycle starting points demonstrated comparable ROCK
potency to piperidine 2 with a preference for the R-enantiomers of
both the piperidine 10 and pyrrolidine 11 as previously reported
(Table 2).7
a-Phenyl substitution of either the piperidine or pyrrol-
idine led to a loss in potency (compare 9–12 and 11–13). However,
both the racemic trans b-substituted piperidine 14 and pyrrolidine
15 improved molecular potency which also translated into
Figure 3. Overlay of inhibitor 31 (blue) and 32 (salmon) in a ROCK homology
model.

T. Bosanac et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3746–3749
3749
Figure 4. X-ray co-crystal structure of 31 bound in ROCK1. The section of the G-loop in the co-structure with inhibitor 22 from Ref. 7 is shown in magenta to highlight the
movement of the G-loop caused by the phenyl ring of 31.
6. Wu, F.; Büttner, F. H.; Chen, R.; Hickey, E.; Jakes, S.; Kaplita, P.; Kashem, M. A.;
References and notes
Kerr, S.; Kugler, S.; Paw, Z.; Prokopowicz, A.; Shih, C.-K.; Snow, R.; Young, E.;
Cywin, C. Bioorg. Med. Chem. Lett. 2010, this issue.
1. (a) Wettschureck, N.; Offermanns, S. J. Mol. Med. 2002, 80, 629; (b) Amano, M.;
7. Ginn, J. D.; Bosanac, T.; Chen, R.; Cywin, C.; Hickery, E.; Kashem, M.; Kerr, S.;
Fukata, Y.; Kaibuchi, K. Exp. Cell Res. 2000, 261, 44; (c) Riento, K.; Ridley, A. J.
Kugler, S.; Prokopowicz III, A.; Schlyer, S.; Smith, J. D.; Turner, M. R.; Wu, F.;
Nat. Rev. Mol. Cell Biol. 2003, 4, 446. Two isoforms of Rho-kinase (ROCK1 and
Young, E. R. R. Bioorg. Med. Chem. Lett. 2010, this issue.
ROCK2) are known and share >90% homology in the kinase domain. No
8. Isonipocotic acids were commercially available. Pyrrolidine acids were either
significant differences in the SAR of the isoquinolone series between the two
commercially available or prepared by a [3+2] cycloaddition (a) Padwa, A.;
isoforms observed.
Dent, W. J. Org. Chem. 1987, 52, 235; (b) Ujjainwalla, F.; Warner, D.; Snedden,
2. (a) Stavenger, R. A. Annu. Rep. Med. Chem. 2008, 43, 87. and references therein;
K.; Grisson, R. D.; Walsh, T. F.; Wyvratt, M. J.; Kalyani, R. N.; MacNeil, T.; Tang,
(b) Oka, M.; Fagan, K. A.; Jones, P. L.; McMurtry, I. F. Br. J. Pharmacol. 2008, 155,
R.; Weinberg, D. H.; Van der Ploeg, L.; Goulet, M. T. Bioorg. Med. Chem. Lett.
444; (c) Schroeter, T.; Griffin, E.; Weiser, A.; Feng, Y.; LoGrasso, P. Biochem.
2005, 15, 4023.
Biophys. Res. Commun. 2008, 374, 356.
9. Doe, C.; Bentley, R.; Behm, D. J.; Lafferty, R.; Stavenger, R.; Jung, D.; Bamford,
3. (a) Sugimoto, M.; Nakayama, M.; Goto, T. M.; Amano, M.; Komori, K.; Kaibuchi,
M.; Panchal, T.; Grygielko, E.; Wright, L. L.; Smith, G. K.; Chen, Z.; Webb, C.;
K. Biochem. Biophys. Res. Commun. 2007, 361, 462; (b) Nohria, A.; Grunert, M. E.;
Khandekar, S.; Yi, T.; Kirkpatrick, R.; Dul, E.; Jolivette, L.; Marino, J. P., Jr.;
Rikitake, Y.; Noma, K.; Prsic, A.; Ganz, P.; Liao, J. K.; Creager, M. A. Circ. Res.
Willette, R.; Lee, D.; Hu, E. J. Pharmacol. Exp. Ther. 2007, 320, 89.
2006, 99, 1426.
10. Rat PK data was obtained from an IV cassette study (n = 3, 0.2 mg/kg) followed
4. (a) Simoes, R. L.; Fierro, I. M. J. Immunol. 2005, 175, 1843; (b) Bardi, G.; Niggli,
by an oral leg (n = 1, 10 mg/kg) in male Sprague-Dawley rats.
V.; Loetscher, P. FEBS Lett. 2003, 542, 79; (c) Niggli, V. FEBS Lett. 1999, 445, 69.
11. Samuel, K.; Yin, W.; Stearns, R. A.; Tang, Y. S.; Chaudhary, A. G.; Jewell, J. P.;
5. (a) Koshikawa, S.; Nishikimi, T.; Inaba, C.; Akimoto, K.; Matsuoka, H. J.
Lanza, T., Jr.; Lin, L. S.; Hagmann, W. K.; Evans, D. C.; Kumar, S. J. Mass Specrom.
Hypertens. 2008, 26, 1837; (b) Katoh, K.; Kano, Y.; Ookawara, S. Trends Cell
2003, 38, 211.
Mol. Biol. 2005, 1, 83; (c) Harvey, K. A.; Paranavitana, C. N.; Zaloga, G. P.;
12. Compound 32 was tested against the same panel of kinases in Ref. 7 (IC50 (nM);
Siddiqui, R. A. J. Cell. Physiol. 2007, 211, 353; (d) Amano, M.; Chihara, K.; Kimura,
PRKG2 = 24, PRKCL2 = 14, PRKCE = 1000, CDC42 = 54).
K.; Fukata, Y.; Nakamura, N.; Matsuura, Y.; Kaibuchi, K. Science 1997, 275, 1308;
(e) Katoh, K.; Kano, Y.; Ookawara, S. Genes Cells 2007, 12, 623.