ROCK inhibitors 2. Improving potency, selectivity and solubility through the application of rationally designed solubilizing groups

Article abstract of DOI:10.1016/j.bmcl.2018.06.043

Solubilizing groups have been frequently appended to kinase inhibitor drug molecules when solubility is insufficient for pharmaceutical development. Such groups are usually located at substitution sites that have minimal impact on target activity. In this report we describe the incorporation of solubilizing groups in a class of Rho kinase (ROCK) inhibitors that not only confer improved solubility, but also enhance target potency and selectivity against a closely related kinase, PKA.

Full text of DOI:10.1016/j.bmcl.2018.06.043

Accepted Manuscript
ROCK Inhibitors 2. Improving Potency, Selectivity and Solubility through the
Application of Rationally Designed Solubilizing Groups
Huai Gao, Craig Marhefka, Marc D. Jacobs, Jingrong Cao, Upul K. Bandarage,
Jeremy Green
PII:
S0960-894X(18)30532-8
https://doi.org/10.1016/j.bmcl.2018.06.043
BMCL 25924
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
30 May 2018
16 June 2018
21 June 2018
Please cite this article as: Gao, H., Marhefka, C., Jacobs, M.D., Cao, J., Bandarage, U.K., Green, J., ROCK Inhibitors
2. Improving Potency, Selectivity and Solubility through the Application of Rationally Designed Solubilizing
Groups, Bioorganic & Medicinal Chemistry Letters (2018), doi: https://doi.org/10.1016/j.bmcl.2018.06.043
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 2. Improving Potency, Selectivity and Solubility
through the Application of Rationally Designed Solubilizing Groups.
Huai Gao, Craig Marhefka, Marc D. Jacobs, Jingrong Cao, Upul K. Bandarage, Jeremy
Green*
Vertex Pharmaceuticals, Incorporated, 50 Northern Avenue, Boston, MA 02210, USA
* Corresponding author. Telephone: 617-341-6315; email: jeremy_green@vrtx.com
Abstract
Solubilizing groups have been frequently appended to kinase inhibitor drug molecules
when solubility is insufficient for pharmaceutical development. Such groups are usually
located at substitution sites that have minimal impact on target activity. In this report we
describe the incorporation of solubilizing groups in a class of Rho kinase (ROCK)
inhibitors that not only confer improved solubility, but also enhance target potency and
selectivity against a closely related kinase, PKA.
Keywords
solubilizing group, structure-based design, rho kinase, ROCK
Page 1 of 14

Graphical Abstract
Page 2 of 14

In the period since the introduction of imatinib (1) in 2001 as the first explicitly designed
therapeutic kinase inhibitor, more than 30 kinase inhibitor drugs have since been
approved by the US FDA. Of these kinase-targeted drugs, 19 contain chemical moieties
specifically incorporated for the purpose of enhancing solubility, since the parent
structure is lacking sufficient solubility for pharmaceutical development (Figure 1: for a
complete illustration of the current FDA-approved kinase inhibitor drugs see Figure S1 in
the Supplementary file). These solubilizing groups are typically attached to a substitution
point on the parent molecule that possesses a high degree of SAR tolerance and usually
neither enhances nor interferes with target potency. The binding implications of these
solubilizing features have been well described and illustrated in a recent review by Wu et
al.1 For example, the X-ray crystal structure of gefitinib (2) (PDB: 2ITY) bound to its
target, EGFR, shows the solubilizing morpholinopropoxy group to be extended away
from the protein surface, not making contact with the protein (Figure 2). Furthermore,
gefitinib (2) reportedly inhibits EGFR with IC50 = 9 nM, while the analogous 6-methoxy
compound has IC50 = 23 nM,2 indicating that the solubilizing group does not significantly
contribute to binding. Interestingly, the X-ray structure of imatinib (1) bound to its target
Abl (PDB: 1IEP) shows the solubilizing methylpiperazine to be in contact with protein
backbone residues (Figure 3). The binding data shows that Bcr-Abl inhibition by imatinib
(1: IC50 = 38 nM) is approximately 5-fold improved over the unsubstituted (4-methyl)
compound (IC50 = 200 nM).3 However, with the exception of the related Bcr-Abl
inhibitors nilotinib and ponatinib, the solubilizing appendages of kinase-inhibitor drugs
extend into solvent space, not providing any additional value in terms of binding
interactions as illustrated by Wu et al.1
Page 3 of 14

Figure 1. FDA approved kinase inhibitors containing solubilizing features (highlighted in
red).
Page 4 of 14

Figure 2. A. Gefitinib (2: R = morpholinopropyl); B. EGFR inhibitory activity of 2
compared with the non-solubilized parent; C. X-ray structure of gefitinb (PDB: 2ITY)
showing the solubilizing morpholinopropyl group extending away from protein surface.
A
C
B
Figure 3. A: Imatinib (1: R = 4-methylpiperazinyl); B. Bcr-Abl inhibitory activity of 1
compared with non-solubilized parent; C. X-ray structure of imatinib (PDB: 1IEP)
showing the methylpiperazinyl group in close proximity (2.7-3.2 Å) to the backbone
carbonyls of Ile360 and His361.
A
C
B
The Rho-associated kinases, ROCK1 and ROCK2 are highly homologous Ser/Thr
kinases, activated by binding of the small GTPase Rho, that act on a variety of substrates,
many of which are implicated in smooth muscle contractility. ROCK inhibitors are
Page 5 of 14

currently under clinical development for a number of therapeutic applications, most
notably for the treatment of glaucoma. These inhibitors have been summarized in a recent
review, which also thoroughly covers the ROCK inhibitor literature.4 We have recently
reported our approach to the discovery and design of pyridyl-thiazole and pyridyl-
thiophene ROCK inhibitors, exemplified by compounds 20 and 21 shown in Figure 4.5
X-ray crystallography and molecular modeling facilitated the optimization of this series
of molecules from the early lead (20) to an optimized molecule (21) possessing properties
suitable for pharmacological evaluation following oral dosing. During the course of this
work key challenges included enhancing potency and selectivity, and improving
solubility.
Figure 4.
Figure 5 illustrates the X-ray structure of compound 20 bound to ROCK. Notable
interactions include hydrogen bonds between the pyridine nitrogen and Met156 (2.9 Å),
the amide carbonyl and the sidechain of catalytic Lys105 (2.8 Å), and the methoxy
oxygen and the backbone NH of Phe87 (3.3 Å). In addition, we noted that Asp117, some
12 Å distant from the methoxy group, might offer an additional binding opportunity. Not
only could this residue engage in an ionic interaction with a pendant basic moiety, such
Page 6 of 14

as piperidine or piperazine, but the analogous residue in PKA, a closely related AGC-
family kinase6 and a key anti-target, is Gln84. The greater steric requirements and neutral
nature of this side chain should enhance selectivity for ROCK. To access this region of
the target, we prepared 3-substituted phenylacetic acid derivatives bearing extended
solubilizing groups and incorporated these into ROCK inhibitors, as illustrated in
Schemes 1 and 2.
Figure 5. X-ray structure of 20 bound to ROCK1 (PDB ID: 4YVE). The Gly-rich loop is
shown in orange and the distal outer loop is shown in blue.
Page 7 of 14

Scheme 1. Preparation of compounds 22 – 29.
(a) K2CO3, acetone, reflux; (b) 1N NaOH, dioxane; (c) Ms-Bt, Et3N, THF, reflux; (d)
DMSO, 70 °C
Scheme 2. Preparation of compounds 30 and 31.
(a) DIAD, PPh3, THF; (b) 2N NaOH, MeOH; (c) Ms-Bt, Et3N, THF, reflux; (d) TFA,
CH2Cl2; (e) 37% HCHO, HCOOH, MeOH, reflux
N-Piperidine, -piperazine and -morpholine compounds 22-29 were prepared as shown in
Scheme 1. Methyl 3-hydroxyphenylacetic acid was alkylated with 2-chloro-1-bromo-
ethane, or 3-chloro-1-bromopropane, under basic conditions in acetone at reflux for 24 h.
Subsequent base hydrolysis yielded the phenylacetic acids, which were coupled to 4-
Page 8 of 14

(pyridin-4-yl)thiazol-2-amine using 1-methanesulfonylbenzotriazole (Ms-Bt),7 as
previously described.5 Conversion to the final products was accomplished by treatment
with the appropriate secondary amine in DMSO at 60-90°C. Products were purified by
preparative HPLC. The preparation of 4-substituted piperidine compounds 30 and 31 is
shown in Scheme 2. Methyl 3-hydroxyphenylacetic acid was alkylated with N-Boc-4-(3-
hydroxypropyl)piperidine under Mitsunobu conditions, the product hydrolyzed and
coupled as described above. Boc deprotection using TFA gave compound 30, while
Eschweiler-Clarke methylation gave compound 31.
Evaluation of the effects of the solubilizing groups is summarized in Table 1.
Page 9 of 14

Table 1. ROCK1 and PKA inhibition by solubilized ROCK inhibitors.
ROCK1
Ki, nMa
PKA Ki,
nMa
Solubility,
µMb
Compd
20
n
Y
0
Me
83c
1400c
22
22
23
24
25
26
27
28
29
30
31
2
2
2
2
3
3
3
3
3
3
510
580
>200
430
47
>4000
>4000
>4000
>4000
>4000
>4000
>4000
>4000
>4000
>200
>200
>200
>200
>200
>200
>200
>200
>200
88
220
240
51
45
13
10
a Enzyme inhibition assays were performed as previously described5
b Solubility measured in 10 mM phosphate buffer (pH 7.4)
c Ki data generated contemporaneously with compounds 22 to 31 and differs slightly
from previously reported data5
The introduction of a solubilizing side-chain, attached at the 3-position of the
phenylacetamide led to compounds with enhanced potency relative to the parent
molecule 20. With a 2-carbon linker, the piperidine (22) and morpholine (23) derivatives
Page 10 of 14

were ca. 5- to 6-fold less potent than 20, indicating that neither occupancy of the above-
described space, nor the presence of a basic, solubilizing center offers any particular
advantage to ROCK inhibition. The analogous piperazine derivatives 24 and 25 were
comparable to 20 in inhibitory potency, suggesting that activity could be restored, though
not improved relative to 20. Compounds 24 and 25 showed superior selectivity against
PKA relative to 20. Extending the linker to 3 carbon atoms in compounds 26 to 29
generally shows enhanced target potency, relative to the 2-carbon linked analogs. Thus
both piperidine (26) and morpholine (27) derivatives are approximately 2- to 3-fold
improved relative to the 2-carbon analogs, but not improved compared to 20. Again this
supports the hypothesis that space occupancy and basicity are neither advantageous, nor
particularly detrimental. However, compounds 28 to 29, in which a basic nitrogen atom
exists at the terminus of the molecule show modestly enhanced potency relative to the
parent molecule 20. In particular, the most basic compounds, the piperidine derivatives
30 and 31 exhibit the greatest enhancement in potency, approximately 8-fold relative to
20. These results suggest that the size of the substituent and the presence of the remote
basic center contribute to the improved inhibitory activity. In all cases activity towards
PKA remains >4 µM, again suggesting that the enhanced activity towards ROCK is due
to specific interactions that are not achieved with a closely related kinase. In addition all
molecules bearing a solubilizing group show enhanced solubility (> 10-fold) relative to
20 in pH 7.4 phosphate buffer (Table 1).
The X-ray crystal structure of 30 bound to ROCK supports our hypothesis. As shown in
Figure 6, the piperidine nitrogen of 30 is in close proximity (2.9 Å) with the side chain of
Asp117, while the rest of the molecule binds in the same conformation observed with
Page 11 of 14

compound 20 and, in general, makes the same contacts with the enzyme as described
above. However the Gly-rich loop has moved upwards (as illustrated), possibly to
accommodate the carbon chain of the solubilizing group, resulting in the apparent loss of
an H-bond between the loop and the meta-O atom. This might explain why greater gains
in activity were not observed. Furthermore compound 30 showed no significant inhibition
towards 20 kinases tested (all >1 µM), while compound 28 showed no inhibition of 145
kinases tested, including 20 AGC-family kinases (Tables S2, S3 in Supplementary
Material).
Figure 6. Compound 30 bound to ROCK1 (PDB ID: 5HVU)
In summary, we have illustrated a strategy for the introduction of solubilizing groups that
accomplishes more than mere solubility enhancement. By identifying key residues
proximal to the active site, solubility, enhanced potency and greater target selectivity
were all accomplished by introducing a soluble moiety that provides additional unique
Page 12 of 14

interactions. Through the application of structure-guided design, we have developed a
series of potent, selective and soluble inhibitors of ROCK. Modeling and X-ray
crystallographic methods helped to identify a key amino acid (Asp117) that was targeted
to provide an additional binding interaction with inhibitors bearing an appended
solubilizing group. Connection of a solubilizing piperidine or piperazine through a 3-
carbon linker gave potent, selective and soluble inhibitors of ROCK. This work
illustrates a basis for future design of kinase inhibitors bearing solubilizing groups that
accomplish potency and selectivity improvements, in addition to enhanced solubility.
Acknowledgements
The authors wish to thank Cameron Stuver-Moody for enzyme inhibition data, and
Georgia McGaughey, Brian Goldman and Daniel McMasters for critical reading of the
manuscript.
References
1. Wu, P.; Nielsen, T. E.; Clausen, M. H. Trends Pharmacol. Sci. 2015, 36, 422–439.
2. Barker, A. J.; Gibson, K. H.; Grundy, W.; Godfrey, A. A.; Barlow, J. J.; Healy, M. P.;
Woodburn, J. R.; Ashton, S. E.; Curry, B. J.; Scarlett, L.; Henthorn, L.; Richards, L.
Bioorg. Med. Chem. Lett. 2001, 11, 1911–1914.
3. Zimmermann, J.; Buchdunger, E.; Mett, H.; Meyer, T.; Lydon, N. B. Bioorg. Med. Chem.
Lett. 1997, 7, 187–192.
4. Feng, Y.; LoGrasso, P. V.; Defert, O.; Li, R. J. Med. Chem. 2016, 59, 2269–2300.
5. 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,
5028–5037.
6. Hanks, S. K. Genome Biol 2003, 4, 111.
7. Katritzky, A. R.; He, H.-Y.; Suzuki, K. J. Org. Chem. 2000, 65, 8210–8213.
Page 13 of 14

ROCK Inhibitors 2. Improving Potency, Selectivity and Solubility
through the Application of Rationally Designed Solubilizing Groups.
Huai Gao, Craig Marhefka, Marc D. Jacobs, Jingrong Cao, Upul K. Bandarage, Jeremy
Green*
Vertex Pharmaceuticals, Incorporated, 50 Northern Avenue, Boston, MA 02210, USA
Highlights
 Solubilizing groups are used to enhance solubility, potency and selectivity
 Structure-based design used to target key interactions with solubilizing groups
 Approximately half of FDA-approved kinase inhibitor drugs contain a
solubilizing group
Page 14 of 14