ROCK inhibitors 3: Design, synthesis and structure-activity relationships of 7-azaindole-based Rho kinase (ROCK) inhibitors

Article abstract of DOI:10.1016/j.bmcl.2018.06.040

Rho kinase (ROCK) inhibitors are potential therapeutic agents for the treatment of a variety of disorders including hypertension, glaucoma and erectile dysfunction. Here we disclose a series of potent and selective ROCK inhibitors based on a substituted 7-azaindole scaffold. Substitution of the 3-position of 7-azaindole led to compounds such as 37, which possess excellent ROCK inhibitory potency and high selectivity against the closely related kinase PKA.

Full text of DOI:10.1016/j.bmcl.2018.06.040

Accepted Manuscript
ROCK Inhibitors 3: Design, Synthesis and Structure-Activity Relationships of
7
-Azaindole-Based Rho kinase (ROCK) Inhibitors
Upul K. Bandarage, Jingrong Cao, Jon H. Come, John J. Court, Huai Gao, Marc
D. Jacobs, Craig Marhefka, Suganthi Nanthakumar, Jeremy Green
PII:
S0960-894X(18)30529-8
https://doi.org/10.1016/j.bmcl.2018.06.040
BMCL 25921
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
30 May 2018
16 June 2018
18 June 2018
Please cite this article as: Bandarage, U.K., Cao, J., Come, J.H., Court, J.J., Gao, H., Jacobs, M.D., Marhefka, C.,
Nanthakumar, S., Green, J., ROCK Inhibitors 3: Design, Synthesis and Structure-Activity Relationships of 7-
Azaindole-Based Rho kinase (ROCK) Inhibitors, Bioorganic & Medicinal Chemistry Letters (2018), doi: https://
doi.org/10.1016/j.bmcl.2018.06.040
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ROCK Inhibitors 3: Design, Synthesis and Structure-Activity
Relationships of 7-Azaindole-Based Rho kinase (ROCK) Inhibitors
Upul K. Bandarage*, Jingrong Cao, Jon H. Come, John J. Court, Huai Gao, Marc D.
Jacobs, Craig Marhefka, Suganthi Nanthakumar, and Jeremy Green*
Vertex Pharmaceuticals Incorporated, 50 Northern Avenue, Boston, MA 02210, USA
*
Corresponding authors.
UB telephone: 617-341-6882; email: upul_bandarage@vrtx.com;
JG telephone: 617-341-6315; email: jeremy_green@vrtx.com
_____________________________________________________________
Abstract
Rho kinase (ROCK) inhibitors are potential therapeutic agents for the treatment of a
variety of disorders including hypertension, glaucoma and erectile dysfunction. Here we
disclose a series of potent and selective ROCK inhibitors based on a substituted 7-
azaindole scaffold. Substitution of the 3-position of 7-azaindole led to compounds such
as 37, which possess excellent ROCK inhibitory potency and high selectivity against the
closely related kinase PKA.
Keywords
7
-azaindole, solubilizing group, structure-based design, lipophilic efficiency, rho kinase,
ROCK
Page 1 of 19

Graphical Abstract
Page 2 of 19

The Rho kinases (ROCK) are Ser/Thr kinases of the AGC-family involved in the regula-
tion of vital cellular processes such as motility, morphology, division, differentiation and
1
contraction. Two kinase isoforms, ROCK-1 and ROCK-2, have been identified that
share 65% overall sequence homology and 92% sequence identity in their ATP-binding
2
domain. Activation of ROCK by GTP-bound Rho leads to phosphorylation of various
protein substrates, including most notably myosin light chain (MLC) and myosin
3
phosphatase -both regulators of smooth muscle contraction. Thus, ROCK inhibitors may
4
have clinical potential as treatments for a variety of disorders such as hypertension,
5
6
glaucoma and erectile dysfunction. To date, three ROCK inhibitors have been approved
7
for clinical use: fasudil (1), approved for the treatment of cerebral vasospasm in Japan
8
and China; ripasudil (2) approved in Japan for the treatment of glaucoma; and more
recently, a third ROCK inhibitor, netarsudil (3), has been approved for the treatment of
elevated intra-ocular pressure in patients with open-angle glaucoma or ocular hyper-
9
tension.
Figure 1. ROCK inhibitors
Page 3 of 19

1
0
In addition to these compounds, many classes of ROCK inhibitor have been reported,
including some containing a 7-azaindole moiety (Figure 1: compounds 4 to 7). LoGrasso
et al. reported azaindole-based inhibitors of ROCK such as 4, however in these inhibitors
1
1,12
the kinase hinge-binding element is the pyrazole group.
Uehata et al. reported the first
valuable ROCK inhibitor tool molecules, including the 7-azaindole derivatives 5
1
3
14
(
Y-30141) and 6 (Y-39983) in the original and subsequent reports. Shirock et al.
reported a series of 3-methyl-4-substituted 7-azaindole inhibitors of ROCK, represented
1
5,16
by compound 7.
Notable among these latter, hinge-binding 7-azaindoles, is the
common 4-substitution. Herein, we report SAR evolution of our previously reported 4-
pyridyl ROCK inhibitors to 3-substituted 7-azaindoles: this non-obvious substitution was
guided by structure-based design.
We previously reported compound 8 in the development of a series of 4-pyridyl-based
1
7
ROCK inhibitors. To further improve our ROCK inhibitors we focused on optimizing
physicochemical properties and reducing the CYP inhibition liability observed within this
series. Although the 2-aminopyridine 8 showed good ROCK inhibition (K
i
= 165 nM),
and 11-fold selectivity against PKA (a closely homologous AGC-family kinase) this
1
7
compound was less potent and selective relative to other 2-substituted pyridines. The
X-ray crystal structure of compound 8 bound to ROCK-1 shows a similar binding
1
7
conformation as previously reported for the 2-unsubstituted system. Notably, hydrogen
bonds exist between the pyridine ring nitrogen 2-amino group and the backbone NH and
CO of Met156 respectively, between the amide carbonyl and the side chain of Lys105,
and between the methoxy group and the NH of both Ala86 and Phe87 of the Gly-rich
loop (Figure 2). It should also be noted that the pyridine and thiazole rings are in a co-
Page 4 of 19

planar arrangement.
Figure 2. Compound 8 bound to ROCK-1 (PDB ID: 5UZJ)
To further develop this series of ROCK inhibitors, we were interested in evaluating
analogous compounds, in which the 2-amino substituent of 8 is connected to the pyridine
ring through a two-carbon (vinyl) spacer forming a 4-substituted 7-azaindole ring system,
as illustrated in Figure 3.
Figure 3. Ring constraint to 4-substituted 7-azaindole system
However, conformational analysis of this 2-dimensional design showed that the desired
coplanarity of the azaindole and thiazole rings would likely be less energetically
Page 5 of 19

favorable. The torsional strain around the biaryl bond of this system is illustrated in
Figure 4A.
Figure 4. Torsion drive of biaryl systems illustrating relative energy (kcal/mol) plotted
1
8
against angle of rotation.
A. 4-Thiazolyl-4-(7-azaindole)
Page 6 of 19

B. 4-Thiazolyl-3-(7-azaindole)
The lowest energy state for the 4-thiazol-4-yl-7-azaindoles occurs when the two ring
systems are at approximately 45° - poorly aligned for optimal kinase active site binding.
Conformations in which the azaindole and thiazole rings are co-planar (0° and 180°) are
at high relative energies. However, when the biaryl connection is shifted to the 3-position
of 7-azaindole, the lowest energy conformation occurs when the thiazole and azaindole
rings are coplanar, with the azaindole pyridine ring adjacent to the thiazole nitrogen
(
Figure 4B, 180°). The opposite coplanar conformation (0° as illustrated in Figure 4B), is
at a higher, though not maximal, energy which occurs when the ring systems are
perpendicular. With the understanding that this structure possesses the important low-
energy coplanarity of the two ring systems, the question remained whether a 3-substituted
7
-azaindole compound would still form effective hydrogen bond interactions with the
hinge region of ROCK, while maintaining the other important binding interactions
Page 7 of 19

between inhibitor and enzyme. Molecular modeling of the 3-substituted-7-azaindole
system docked in the ROCK active site suggested that, despite the shift of substitution
from the pyridyl ring (4 position substituted azaindole) to the pyrrole ring (3 position
substituted azaindole), the relative location of the pyridine nitrogen would still support
hydrogen bonding with Met156. Thus we hypothesized that the 7-azaindole system might
provide an alternative hinge-binder to replace 2-aminopyridine and, counter intuitively,
that the 7-azaindole 3-position might provide both a suitable vector to direct substituents,
as well as an appropriate 3-dimensional configuration. Contacts with the enzyme that had
previously been exploited were expected to be retained. To test this hypothesis,
compounds were prepared bearing the 2-(phenylacetylamino)-thiazole substituent at both
the 3- and 4-position on the 7-azaindole ring.
The synthesis of 4-substituted azaindole compounds 9 and 13 was achieved in six steps,
as shown in Scheme 1. Benzenesulfonyl protection of 4-chloro-7-azaindole 10, followed
by reaction with 1-ethoxyvinyl tri-n-butyltin in the presence of PdCl (PPh ) yielded enol-
2
3 2
ether 11 in moderate yield. Reaction of 11 with NBS afforded -bromoketone 12, which
was reacted with thiourea, to afford the aminothiazole ring system. Subsequent acylation
and followed by acidolytic removal of the benzenesulfonyl protecting group afforded
compounds 9 and 13.
Page 8 of 19

Scheme 1. Reagents and conditions: (a) NaH, benzenesulfonyl chloride, THF, RT; (b) 1-
ethoxyvinyl tri-n-butyltin, PdCl (PPh ) , dioxane, 110 °C; (c) NBS, H O, THF, RT; (d)
2
3
2
2
thiourea, EtOH, reflux; (e) phenylacetyl chloride or 3-methoxyphenylacetyl chloride,
Et N, DCM, RT; (f) 6N HCl, MeOH, reflux.
3
3
-Substituted-7-azaindoles (17-32) were prepared in three steps as shown in Scheme 2.
Friedel-Crafts acylation of commercially available 7-azaindole 14 with bromoacetyl
bromide in dichloromethane afforded 3-(2-bromoacetyl)-7-azaindole 15 in 64% yield.
Hantzsch thiazole synthesis using thiourea in refluxing EtOH the thiazole 16 in 83%
yield. Amidation of compound 16 with various m-substituted phenylacetic acids was
1
7
performed as previously described using methanesulfonylbenzotriazole (MeSO
2
Bt)
reagent under microwave heating at 160°C for 10 minutes to afford final products 17 to
2 in good yields.
1
9
3
Page 9 of 19

Scheme 2. Reagents and conditions: (a) BrCH COBr, AlCl , DCM, 50 °C; (b) thiourea,
2
3
EtOH, reflux; (c) phenylacetic acids, MeSO Bt, Et N, THF, DMA, 160 °C.
2
3
Enzyme inhibition data for compounds is summarized in Table 1, which shows that the 7-
azaindole analogs are strong inhibitors of ROCK-1, with K values typically below
i
2
0 nM. However, most of these azaindole derivatives showed reduced selectivity against
PKA. In particular, the 4-substituted compounds 9 and 13 showed the least selectivity,
2
0
and somewhat weaker ROCK affinity, resulting in lower LLE values. In the analogous
-substituted 7-azaindole series, the overall affinity for ROCK was slightly improved
hence improved LLE values), though selectivity against PKA, in most cases, was not
3
(
markedly better. Interestingly, the sulfonamides (23 to 32) showed some selectivity
against PKA with the sulfonamide oriented in either direction. The 3-(N-sulfonylamino)-
phenylacetyl compounds 28 to 32 showed the highest selectivity, particularly with larger
N-substituents. Inhibition of CYP 3A4, an issue previously observed in the pyridine
series, remained stubbornly persistent with these analogous azaindole derivatives (Table
1
).
Page 10 of 19

Table 1. Kinase inhibition, selectivity, ligand efficiency and CYP 3A4 inhibition of
azaindole derivatives
a
Compd. ROCK K
i
, nM PKA K
i
, nM
Selectivity Ratio LLE (ROCK K
PKA K /ROCK K
i
)
CYP 3A4
IC50, µM
26
NA
NA
NA
NA
NA
2.0
(
i
i
)
8
165
19
16
5
3
3
6
14
40
11
19
3
6
2
2
<1
2
1300
9
7.9
0.5
1.4
1.2
1.3
2
4.0
4.1
4.3
4.8
5.2
5.1
4.9
4.3
4.9
5.2
4.9
5.1
5.0
6.2
6.6
6.3
5.6
5.6
5.2
13
9
23
6
4
6
1
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
13
2.2
7.9
10.5
14.5
5.8
110
420
160
110
60
39
31
48
180
148
350
370
2.5
6.4
2.3
9.3
5.7
2.6
3.4
8.3
5.9
3.6
4.9
3.7
20
6.5
15.5
24
>180
74
175
185
2
2
a
Lipophilic ligand efficiency (LLE = pK – clogP)
i
To understand the binding modes of these compounds, the X-ray crystal structure of
compound 28 bound to ROCK (PDB ID: 5KKS) was obtained (Figure 5). As expected
the 7-azaindole moiety of 28 forms a hydrogen bond between the 7-azaindole nitrogen
and Met156. In addition the azaindole pyrrole NH forms a new hydrogen bond
interaction with the backbone carbonyl of Glu154. This is a distinct binding orientation
than that observed for compound 8 (Figure 2), in which the pyridine 2-amino group is
seen to interact with the CO group of Met156: thus an effective 180° rotation has been
achieved, in accord with the models presented above. Notable additional interactions
between 28 and ROCK are the previously observed H-bonding interaction with catalytic
Page 11 of 19

Lys105 and two hydrogen bonds between the sulfonamide and the backbone NH of
residues Phe87 and Ala86.
Figure 5. X-ray crystal structure of compound 28 bound to ROCK (PDB ID: 5KKS).
2
1
In our companion paper we have described how the addition of a pendant solubilizing
group, attached to the meta-position of the phenylacetamide, is able to confer both
improved potency and selectivity due to the additional interactions made between a distal
basic nitrogen and Asp117 of the outer loop. We sought to apply this design concept in
the 7-azaindole series.
Table 2. Kinase inhibition, selectivity, ligand efficiency and CYP 3A4 inhibition of
azaindole derivatives bearing solubilizing groups
Page 12 of 19

ROCK K
i
,
Selectivity Ratio
(PKA K /ROCK K
LLE
CYP 3A4
IC50, µM
Compd
R
i
PKA K , nM
a
nM
i
i
)
i
(ROCK K )
3
3
8
290
36
84
5.3
14.9
34
35
36
37
3
<5
10
<1
253
180
470
570
5.2
4.4
5.1
5.6
17.5
20.8
12.6
19.6
>36
47
>570
From the data shown in Table 2 it can be seen that further SAR development was attained
through the addition of a solubilizing group able to interact with Asp117, resulting in
increased ROCK inhibitory potency (and sustained LLE) and increased selectivity against
PKA. An explanation for the observed selectivity is illustrated by the X-ray
crystallographic structures shown in Figure 6A, where compound 37 is shown bound to
ROCK. The distal piperazine N-Me is seen in close proximity (3.0 Å) to the side chain
carboxylate of Asp117. In Figure 6B is the X-ray structure of the co-crystal of 35
(
yellow) bound to PKA. In this structure, the solubilizing group is folded away from PKA
Gln84, supporting the design hypothesis described above. The combination of the larger
spatial requirement of Gln84 and the absence of ionic interaction, requires that the basic
residue and flexible 3-carbon atom linker must move in order for the rest of the molecule
to bind. A further benefit of the solubilizing group is that CYP 3A4 inhibition was
generally improved relative to both the other azaindole derivatives, and to our previously
1
7
reported pyridine-based inhibitors.
Page 13 of 19

Figure 6. (A) X-ray structure of compound 37 (yellow) bound to ROCK (blue) (PDB ID:
5
5
KKT); (B) X-ray structure of compound 35, (yellow) bound to PKA (green) (PDB ID:
UZK) help explain selectivity.
In conclusion, we have developed a series of highly potent and PKA selective 7-
azaindole-based ROCK inhibitors. In this context, the 3-substituted 7-azaindole is a non-
Page 14 of 19

intuitive isostere for 4-substituted 2-aminopyridine as a result of an alternative target
interaction being created between the inhibitor and the kinase hinge region. Other
existing interactions with the enzyme, established for potency and selectivity in the 4-
pyridine series, are retained by the 3-substituted 7-azaindole series. Furthermore, pendant
solubilizing groups were explored and found to further enhance potency and selectivity
by means of specific designed interactions.
Acknowledgements
The authors wish to thank Cameron Stuver-Moody for enzyme inhibition data, and
Georgia McGaughey and Daniel McMasters for critical reading of the manuscript.
REFERENCES
1
2
3
4
5
6
7
. Godoy, M. A. F. de; Rattan, S. Trends Pharmacol. Sci. 2011, 32, 384–393.
. Mueller, B. K.; Mack, H.; Teusch, N. Nat. Rev. Drug Discov. 2005, 4, 387–398.
. Riento, K.; Ridley, A. J. Nat. Rev. Mol. Cell Biol. 2003, 4, 446–456.
. Budzyn, K.; Marley, P. D.; Sobey, C. G. Trends Pharmacol. Sci. 2006, 27, 97–104.
. Rao, V. P.; Epstein, D. L. BioDrugs 2007, 21, 167–177.
. Sopko, N. A.; Hannan, J. L.; Bivalacqua, T. J. Nat. Rev. Urol. 2014, 11, 622–628.
. Shibuya, M.; Satoh, S.; Suzuki, Y. In Cerebral Vasospasm: Advances in Research and
Treatment; Macdonald, R., Ed.; Thieme: New York, 2004; pp. 284–287.
. Garnock-Jones, K. P. Drugs 2014, 74, 2211–2215.
8
9
. Sturdivant, J. M.; Royalty, S. M.; Lin, C.-W.; Moore, L. A.; Yingling, J. D.; Laethem, C.
L.; Sherman, B.; Heintzelman, G. R.; Kopczynski, C. C.; deLong, M. A. Bioorg. Med.
Chem. Lett. 2016, 26, 2475–2480.
Page 15 of 19

1
1
0. Feng, Y.; LoGrasso, P. V.; Defert, O.; Li, R. J. Med. Chem. 2016, 59, 2269–2300.
1. Chowdhury, S.; Sessions, E. H.; Pocas, J. R.; Grant, W.; Schröter, T.; Lin, L.; Ruiz, C.;
Cameron, M. D.; Schürer, S.; LoGrasso, P.; Bannister, T. D.; Feng, Y. Bioorg. Med.
Chem. Lett. 2011, 21, 7107–7112.
1
2. Sessions, E. H.; Chowdhury, S.; Yin, Y.; Pocas, J. R.; Grant, W.; Schröter, T.; Lin, L.;
Ruiz, C.; Cameron, M. D.; LoGrasso, P.; Bannister, T. D.; Feng, Y. Bioorg. Med. Chem.
Lett. 2011, 21, 7113–7118.
1
1
1
3. Uehata, M.; Ishizaki, T.; Satoh, H.; Ono, T.; Kawahara, T.; Morishita, T.; Tamakawa, H.;
Yamagami, K.; Inui, J.; Maekawa, M.; Narumiya, S. Nature 1997, 389, 990–994.
4. Sagawa, H.; Terasaki, H.; Nakamura, M.; Ichikawa, M.; Yata, T.; Tokita, Y.; Watanabe,
M. Exp. Neurol. 2007, 205, 230–240.
5. 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
2
008, 3, 1893–1904.
6. Schirok, H.; Paulsen, H.; Kroh, W.; Chen, G.; Gao, P. Org. Process Res. Dev. 2009, 14,
68–173.
1
1
1
7. Green, J.; Cao, J.; Bandarage, U. K.; Gao, H.; Court, J.; Marhefka, C.; Jacobs, M.;
Taslimi, P.; Newsome, D.; Nakayama, T.; Shah, S.; Rodems, S. J. Med. Chem. 2015, 58,
5
028–5037.
1
8. Torsion scanning was conducted using OpenEye Scientific Software employing the
MMFF94 forcefield.
1
2
9. Katritzky, A. R.; He, H.-Y.; Suzuki, K. J. Org. Chem. 2000, 65, 8210–8213.
0. Leeson, P. D.; Springthorpe, B. Nat. Rev. Drug Discov. 2007, 6, 881–890.
Page 16 of 19

2
1. Huai Gao; Craig Marhefka; Marc Jacobs; Cao, J.; Bandarage, U. K.; Green, J. Bioorg.
Med. Chem. Lett. 2017.
Page 17 of 19

ROCK Inhibitors 3: Design, Synthesis and Structure-Activity
Relationships of 7-Azaindole-Based Rho kinase (ROCK) Inhibitors
Upul K. Bandarage*, Jingrong Cao, Jon H. Come, John J. Court, Huai Gao, Marc D.
Jacobs, Craig Marhefka, Suganthi Nanthakumar, and Jeremy Green*
Vertex Pharmaceuticals, Incorporated, 50 Northern Avenue, Boston, MA 02210, USA
Highlights



3-Substituted 7-azaindoles are shown to be bioisosteric replacements for 4-substituted pyridines
Solubilizing groups are used to enhance solubility, potency and selectivity
X-ray crystallography of target-ligand complexes provides clear illustration of selectivity
Page 18 of 19

ROCK Inhibitors 3: Design, Synthesis and Structure-Activity
Relationships of 7-Azaindole-Based Rho kinase (ROCK) Inhibitors
Upul K. Bandarage*, Jingrong Cao, Jon H. Come, John J. Court, Huai Gao, Marc D.
Jacobs, Craig Marhefka, Suganthi Nanthakumar, and Jeremy Green*
Vertex Pharmaceuticals Incorporated, 50 Northern Avenue, Boston, MA 02210, USA
Graphical Abstract
Page 19 of 19