A three-step protocol for lead optimization: Quick identification of key conformational features and functional groups in the SAR studies of non-ATP competitive MK2 (MAPKAPK2) inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.11.074

A three-step protocol for SAR development was introduced and applied to the SAR studies of the MK2 inhibitor program. Following this protocol, key conformational features and functional groups for improving MK2 inhibitor activity were quickly identified.

Full text of DOI:10.1016/j.bmcl.2011.11.074

Bioorganic & Medicinal Chemistry Letters 22 (2012) 65–70
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
A three-step protocol for lead optimization: Quick identification of key
conformational features and functional groups in the SAR studies
of non-ATP competitive MK2 (MAPKAPK2) inhibitors
Xianhai Huang a, , Xiaohong Zhu a, Xiao Chen a, Wei Zhou a, Dong Xiao a, Sylvia Degrado a,
⇑
Robert Aslanian a, James Fossetta b, Daniel Lundell b, Fang Tian b, Prashant Trivedi b, Anandan Palani a
a Department of Medicinal Chemistry, Merck Research Laboratories, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
b Department of In Vitro Biology, Merck Research Laboratories, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A three-step protocol for SAR development was introduced and applied to the SAR studies of the MK2
inhibitor program. Following this protocol, key conformational features and functional groups for
improving MK2 inhibitor activity were quickly identified. Through this effort, the initial gap observed
between in vitro binding activity and cellular activity in the lead identification stage was very much
reduced. Compound 28 was identified with single digit binding activity (IC50 = 8 nM) and good cellular
activity (EC50 = 310 nM). This provides further evidence that non-ATP-competitive binding MK2 inhibi-
tors are feasible by targeting the outside ATP pocket.
Received 26 October 2011
Revised 14 November 2011
Accepted 18 November 2011
Available online 23 November 2011
Keywords:
Three-step protocol
SAR
Ó 2011 Elsevier Ltd. All rights reserved.
Conformation
MK2
TNFa
Non-ATP competitive
Inhibitor
In recent years, the pharmaceutical industry has been undergo-
ing a series of dramatic changes including mega scale mergers1 and
extensive outsourcing spree.2 Part of the reasons for these changes
is the need to speed up innovation and improve the efficiency of
drug discovery and development. How to minimize cost, reduce
failure and accelerate the process from target identification to a
drug on the market has become a million dollar question. Lead
optimization (LO) plays important role in this process since every
drug has to come from a development candidate resulting from a
LO effort. The conformation of a drug entity is the most important
aspect when the drug interacts with its target protein/enzyme
to achieve the desired physiological effect. Thus we propose a
three-step protocol for lead optimization based on defining
the conformation of an organic compound. Step 1: define the
intrinsic conformation of the compound by generating the core
structure (the trunk, through first degree SAR studies) which will
establish the groundwork for good interaction with the target; Step
2: define the extrinsic conformation by identifying preferred
orientation of main pieces (the branches, through secondary SAR
studies) to achieve desired activity; Step 3: localized conformation
tuning by working on focused SAR studies at specific pieces of the
molecule (the leaves, through third degree SAR studies) to identify
compounds with optimum PK/PD profiles. We present here the
application of this protocol in the MK2 LO program to showcase
how we can potentially improve the SAR development process.
Protein kinases play very important role in inflammatory dis-
eases.3,4 Although a large number of companies are working very
aggressively in this field, only a few kinase inhibitors have been ap-
proved by the FDA in recent years. One of the reasons for the lim-
ited success is due to the safety concern related with protein kinase
inhibitors. Among these targeted protein kinases, P38 mitogen-
activated protein kinase (MAPK) was evaluated for diseases such
as rheumatoid arthritis (RA), Crohn’s disease, inflammatory bowel
syndrome (IBS), and psoriasis in numerous early stage clinical tri-
als,5–8 however, no p38 MAPK inhibitors have progressed beyond
phase II trials due to dose dependent toxicities and lack of effica-
cies.9,10 To minimize the side effects caused by directly targeting
p38 MAPK, inhibiting the downstream kinases such as p38
MAPK-activated protein kinase 2 (MAPKAPK2, MK2) has becoming
an attractive strategy. MK2 has been implicated in promoting pro-
inflammatory cytokine production and related inflammatory dis-
eases,11,12 and is the primary MAPKAPK family member required
⇑
Corresponding author. Tel.: +1 908 740 3487; fax: +1 908 740 7664.
E-mail addresses: xianhai.huang@merck.com, xianhai.huang@spcorp.com
for lipopolysaccharide (LPS)-induced tumor necrosis factor
a
(TNFa
) and interleukin 6 (IL-6) production.13–15 Studies have
(X. Huang).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.11.074

66
X. Huang et al. / Bioorg. Med. Chem. Lett. 22 (2012) 65–70
NH
NH
N
O
N
O
O
R
N
O
Cl
N
H
N
2
1
2a R = Cl, IC50 = 110 nM
Cellular Data:
pHSP27 EC50 (nM): 350
TNF EC50 (uM) = 5
α
2bR = CN, IC50 = 115 nM
Figure 1. MK2 inhibitor leads from lead identification.
which could offer improved general selectivity and cellular activ-
ity.22 We have recently identified a series of non-ATP competitive
MK2 inhibitors by targeting the outside ATP pocket through the
ALIS screening system and generated lead compound 1 (Fig. 1)
with very good kinase selectivity across the board. Initial hit to
lead effort resulted in compound 2 with improved in vitro bind-
ing activity and cellular potency, which formed the base for the
lead optimization program.22 Compound displayed good cellular
potency in a pHSP27 assay, however, in a more relevant cellular
branch 2
NH
branch 1
core
O
N
O
R
N
N
branch 3
assay (hTNF
active (EC50 = 5
a
assay in THP1 cells) the compound was much less
M). There is a clear gap between in vitro binding
2
l
activity and cellular activity. In order to minimize this gap and
improve the cellular activity, we decided to focus on improving
in vitro binding activity first (IC50 < 50 nM) in hope that the cellu-
lar activity will improve with it (EC50 < 500 nM). Obviously, the
concern is whether we can drive the binding affinity low
enough through the non-competitive mode (allosteric interac-
tion). We report herein how we employed the three-step protocol
discussed above to quickly identify key molecular conformational
features and functional groups responsible for the much im-
proved binding activity and cellular activity of a new series of
MK2 inhibitors.
Figure 2. Main conformation features of MK2 hit lead.
shown that MK2 knock-out mice had reduced joint damage in a
collagen-induced arthritis (CIA) model and no induction of asthma
in a lung ovalbumin sensitization model16,17 where the mice were
health and void of embryonic lethality observed with p38 MAPK
knock-out mice. This provides evidence that MK2 inhibitors may
provide desirable efficacy without toxic side effects. To this end,
a few research laboratories were involved in the development of
ATP-competitive MK2 inhibitors.18–21 However, the results from
Pfizer have indicated that it is difficult to use an ATP-competitive
approach to inhibit MK2 in humans due to poor biochemical effi-
ciency.22 On the other hand, biochemical efficiency (ratio of bind-
ing affinity to target vs cellular activity) may be increased in
MK2 inhibitors through a non-ATP competitive binding mode
For the first step, we needed to identify the best core structure
for further development. This is the most important stage for a lead
optimization program. If the core structure can not provide opti-
mum interaction with the target, the best result will be difficult
N
NH
N
N
O
O
NH
O
O
O
O
Cl
NC
N
N
NC
N
N
N
N
3
5
4
MK2IC50 = 4743 nM
MK2 IC50 = 10000 nM
MK2 IC50 = 10000 nM
H
N
H
N
NH
N
N
N
N
O
O
O
O
O
NC
N
Cl
O
N
Cl
N
N
N
6
8
7
MK2IC50 = 79 nM
MK2IC50 = Not Active
MK2IC50 = Not Active
Figure 3. SAR studies to define the secondary conformation of branch 2. Data represents an average of multiple determinations (2 or more).

X. Huang et al. / Bioorg. Med. Chem. Lett. 22 (2012) 65–70
NH
67
NH
N
N
NH
1-phenyl-1,4-diazepane
2-phenyloctahydropyrrolo[3,4-c]pyrrole
4-phenylpiperidine
NH
NH
NH
N
H
3-phenylpiperidine
2-phenylpiperidine
3-phenylpyrrolidine
3-phenylazetidine
Figure 4. Replacement of 1-phenylpiperazine at branch 2.
to achieve in down-the-road SAR studies. Core structures are usu-
ally generated from the lead identification (LI) stage, and limited
SAR studies can be explored to select the best core (first degree
SAR studies). In the MK2 project, all initial SAR studies in the LI
program pointed to the 5-substituted 2-furanyl amide (Fig. 2) as
the key unit responsible for potency. So we decided to keep this
core untouched during LO development.
For the second step, we needed to define the extrinsic confor-
mation (main branches) of the lead molecule by exploring the
SAR of each branch (secondary SAR studies). As can be seen from
the structure of compound 2, there are three distinct main func-
tional units, branches 1–3 (Fig. 2). SAR effort in branch 1 indicated
that modifications were not very well tolerated and para-chloro
and cyanophenyl groups proved to be the best substitutions. We
decided to focus the secondary SAR on the modification of
branches 2 and 3.
For branch 2 modification (Fig. 3), we introduced a saturated ring
cyclohexyl group (3) to examine the importance of the phenyl ring,
but its activity dropped dramatically. Compounds 4 with a flexible
chain and 5 without a phenyl linker are also much less active. These
Table 1
SAR studies to define the secondary conformation of branch 3a
NH
N
O
O
NC
N
R
Compound no.
R=
IC50 (nM)
hTNF
6.69
a
EC50
Compound
R=
IC50 (nM)
hTNFa EC50
(l
(
l
M)
M)
F
9
41
14
12
10
11
70
28.1
3.46
15
16
128
110
N
9
N
12
13
1920
1450
17
12
8.4
N
F
F
a
Data represents an average of multiple determinations (2 or more).

68
X. Huang et al. / Bioorg. Med. Chem. Lett. 22 (2012) 65–70
Table 2
Localized SAR studies of branch 3a
NH
N
O
O
NC
N
R
Compound
R=
IC50 (nM)
285
hTNF
a
EC50
Compound no.
R=
IC50 (nM)
43
hTNFa EC50
(l
(l
M)
M)
18
29
MeO
N
N
N
19
20
258
21
30
31
24
52
OMe
N
1.77
N
O
MeO
21
22
8
2.95
1.59
20
12
1.96
NC
NC
32
19
0.734
1.59
OMe
33
MeO
23
12
9
N
OMe
R
34
35
N
N
24
9
1.24
100
N
OMe
N
N
O
NH
NH
N
N
N
O
NC
N
N
N
N
25
14
2.54
36
258
O
N
O
N
N
NC
26
27
0.88
37
177
N
N
27
28
20
8
1.80
0.31
OMe
N
OMe
a
Data represents an average of multiple determinations (2 or more).

X. Huang et al. / Bioorg. Med. Chem. Lett. 22 (2012) 65–70
69
results suggest that the flat conformation and the rigidity of a
phenyl ring are important for MK2 binding activity. To further
define the conformation of branch 2, we rotated the piperazine ring
around the phenyl group (Fig. 3, 6–7). It turned out that the ortho-
and meta-substitutions were not tolerated and para-substitution
gave the best result, demonstrating that para-substitution of the
phenyl group is the optimum conformation at branch 2. The next
step for branch 2 modification was confined to the changes at
the para-substitution (third degree SAR). We have introduced func-
tional groups (Fig. 4) such as 1-phenyl-1,4-diazepane (105 nM),
2-phenyloctahydropyrrolo[3,4-c]pyrrole (95 nM), 4-phenylpiperi-
dine (129 nM), 3-phenylpiperidine (324 nM), 2-phenylpiperidine
and cellular activity. This suggests that heteroatomic substitution
may play a role in this MK2 series and provides direction for further
SAR development. Based on this observation, we introduced polar
groups on the phenyl ring such as a nitrile group (21, 22), and both
compounds gave good binding activity with the meta-cyanophenyl
compound 22 having better cellular activity (EC50 = 1.59 lM).
meta-Methoxy substitution is also acceptable (23). We then intro-
duced pyridine substitution (17, 24, 25), resulting in compounds
with consistently good binding activity. The relatively better cellu-
lar activity of 3-pyridyl (24) than 2-pyridyl (25) further confirms
the observation that meta-heteroatom or substitution (22) plays
important role in improving cellular activity. To further take advan-
tage of this trend, compound 26 was prepared. This compound
shows one of the best cellular activity (EC50 = 880 nM) with good
binding activity. With the introduction of a methoxy group to the
pyridine analogs (27-31), we quickly identified compound 28 with
both very good binding (IC50 = 8 nM) and cellular activity
(EC50 = 310 nM) after a few substitution manipulations changing
the methoxy and pyridyl nitrogen locations.
(5.4
(10
l
M), 3-phenylpyrrolidine (10
lM), and 3-phenylazetidine
lM) at branch 2 while maintaining a 2-pyridylmethyl group
at branch 3. It turned out that 1-phenylpiperazine was one of the
best groups at branch 2 and that the terminal basic amino group
was vital for binding and cellular activity. This effort completed
the identification of conformational preference and the optimum
functional group at branch 2.
We next turned our attention to secondary SAR studies of
branch 3 to define the optimum conformation. Based on the results
from previous SAR,22 we decided to introduce a biarylalkyl group
at branch 3. As summarized in Table 1, by rotating the phenyl sub-
stitution from para- to meta- and ortho-position, we see clear bind-
ing activity preference. The para-substitution (9) gives a better IC50
value than meta-substitution (10), while the ortho-substitution
gives the best activity (9 nM, 11). This is a dramatic improvement
in terms of binding activity and it confirms that a single digit nano-
molar IC50 can be achieved under the non-competitive mode. We
also tested these compounds in cellular studies (inhibition of
To demonstrate the importance of localized SAR studies (third
degree SAR), we have prepared compounds 32 to 35 (Table 2) by
introducing pyridyl biaryl alkyl substitution at branch 3. From
the binding data, it is clear that pyridyl biaryls are tolerated and
give comparable results as the phenyl biaryl series of compounds.
Interestingly, the introduction of a single nitrogen atom in com-
pound 33 resulted in improvement of cell activity (compared to
32), but this observation is not repeated in compound 34 (com-
pared to 28) and in 35 (compared to 26). On the other hand, simple
localized modification at branch 1 with preferred orientation at
branch 3, much reduced binding activity is observed in compound
36 (vs 11) and 37 (vs 24).
hTNFa in THP-1 cells), and gratifyingly, we see the alignment of
the cellular activity and the binding activity with the ortho-substi-
tuted compound bearing the best cellular activity. This conforma-
tional preference is further confirmed by the preparation of
additional compounds (12–14 and 15–17).
The synthesis of the biaryl alkyl MK2 inhibitors is straight for-
ward and summarized in Scheme 1. Commercially available com-
pound 38 was coupled with aniline 39 in good yield in the
presence of EDC/HOBt to give amide 40 which was alkylated with
2-bromobenzyl bromide under basic conditions to give bromide
41. Final compounds (42) could be accessed from 41 by Suzuki
coupling and deprotection with TFA. The pyridyl biaryl compounds
were prepared in a similar fashion.
In summary, we have introduced a three-step protocol for SAR
development and applied it to the SAR studies of a novel series
of MK2 inhibitors. To improve the cellular activity profile of the
hit lead 2, we have chosen to keep the 5-phenyl 2-furanyl amide
core and make changes to the three branches of the core. Through
defining the extrinsic conformation (secondary SAR studies), we
After identifying the preferred extrinsic conformation of branch
3, we next focused our effort on the localized SAR studies (third de-
gree SAR) on the modification of ortho-substituted biaryl substitu-
tions in hope to improve the cellular activity while maintain good
in vitro binding activity. The results are summarized in Table 2. We
first introduced ortho-methyl and ortho-, ortho0-dimethyl substi-
tuted phenyl (18, 19) to introduce further conformation interac-
tion, but the activity dropped quite dramatically. On the other
hand, when we prepared compound 20 with ortho-, ortho0-di-
methyl substituted isooxazole, we saw improved binding activity
NBoc
NBoc
EDC, HOBt,
O
N
O
DIPEA, DMF
N
O
NC
+
OH
O
NC
90%
N
H
H2N
38
40
39
NH
N
NBoc
O
Br
N
O
O
NC
N
1. Suzuki coupling
2. TFA
O
NC
N
Br
NaH, DMF, rt
80%
80%
Br
41
42
Scheme 1. The general synthetic route to biarylalkyl MK2 inhibitors.

70
X. Huang et al. / Bioorg. Med. Chem. Lett. 22 (2012) 65–70
8. Ulfgren, A. K.; Andersson, U.; Engstrom, M.; Klareskog, L.; Maini, R. N.; Taylor, P.
C. Arthritis Rheum. 2000, 43, 2391.
have quickly identified key conformational preference for branch 2
as para-piperazine substituted phenyl and branch 3 as ortho-
substituted biphenyl. Localized SAR studies (third degree SAR) on
branch 3 have helped to improve the binding activity to single digit
nanomolar and cellular activity less than 500 nM and identify 2-(4-
methoxypyridin-3-yl)benzyl (in 28) and 2-(pyrimidin-5-yl)benzyl
(in 26) as two of the preferred functional groups responsible for
improved binding and cellular activity. The importance of these
groups is further confirmed by other series of MK2 inhibitors with
the same core structure which will be reported in due courses.23
9. Schett, G.; Zwerina, J.; Firestein, G. Ann. Rheum. Dis. 2008, 67, 909.
10. Alten, R. E.; Zerbini, C.; Jeka, S.; Irazoque, F.; Khatib, F.; Emery, P.; Bertasso, A.;
Rabbia, M.; Caulfield, J. P. Ann. Rheum. Dis. 2010, 69, 364.
11. Kumar, S.; Boehm, J.; Lee, J. C. Nat. Rev. Drug Disc. 2003, 2, 717.
12. Schindler, J. F.; Monahan, J. B.; Smith, W. G. J. Dent. Res. 2007, 86, 800.
13. Kotlyarov, A.; Neininger, A.; Schubert, C.; Eckert, R.; Birchmeier, C.; Volk, H.-D.;
Gaestel, M. Nat. Cell Biol. 1999, 1, 94.
14. Shi, Y.; Kotlyarov, A.; Laab, K.; Gruber, A. D.; Butt, E.; Marcus, K.; Meyer, H. E.;
Friedrich, A.; Volk, H.-D.; Gaestel, M. Mol. Cell. Biol. 2003, 23, 7732.
15. Ronkina, N.; Kotlyarov, A.; Dittrich-Breiholz, O.; Kracht, M.; Hitti, E.; Milarski,
K.; Askew, R.; Marusic, S.; Lin, L.-L.; Gaestel, M.; Telliez, J.-B. Mol. Cell. Biol. 2007,
27, 170.
16. Hegen, M.; Gaestel, M.; Nickerson-Nutter, C. L.; Lin, L.-L.; Telliez, J.-B. J.
Immunol. 2006, 177, 1913.
Acknowledgments
17. Gorska, M. M.; Liang, Q.; Stafford, S. J.; Goplen, N.; Dharajiya, N.; Guo, L.; Sur, S.;
Gaestel, M.; Alam, R. J. Exp. Med. 2007, 204, 1637.
18. Lopes, L. B.; Flynn, C.; Komalavilas, P.; Panitch, A.; Brophy, C. M.; Seald, B. L.
Biochem. Biophys. Res. Commun. 2009, 382, 535.
19. Anderson, D. R.; Meyers, M. J.; Kurumbail, R. G.; Caspers, N.; Poda, G. I.; Long, S.
A.; Pierce, B. S.; Mahoney, M. W.; Mourey, R. J.; Parikh, M. D. Bioorg. Med. Chem.
Lett. 2009, 19, 4882.
20. Anderson, D. R.; Meyers, M. J.; Kurumbail, R. G.; Caspers, N.; Poda, G. I.; Long, S.
A.; Pierce, B. S.; Mahoney, M. W.; Mourey, R. J. Bioorg. Med. Chem. Lett. 2009, 19,
4878.
21. Velcicky, J.; Feifel, R.; Hawtin, S.; Heng, R.; Huppertz, C.; Koch, G.; Kroemer, M.;
Moebitz, H.; Revesz, L.; Scheufler, C.; Schlapbach, A. Bioorg. Med. Chem. Lett.
2010, 20, 1293–1297.
22. Huang, X.; Shipps, G. W.; Cheng, C. C.; Spacciapoli, P.; Zhang, X.; McCoy, M. A.;
Wyss, D. F.; Yang, X.; Achab, A.; Soucy, K.; Montavon, D. K.; Murphy, D. M.;
Whitehurst, C. E. ACS Med. Chem. Lett. 2011, 2, 632. and references cited there
in.
23. (a) Qin, J.; Dhondi, P.; Huang, X.; Aslanian, R. G.; Fossetta, J. D.; Palani, A. ACS
Med. Chem. Lett., accepted for publication.; (b) Xiao, D.; Palani, A.; Aslanian, R.
G.; Degrado, S.; Huang, X.; Zhou, W.; Sofolarides, M.; Chen, X. Patent WO
2011075375 A1, 2011.
We thank Drs. Christopher Boyce and Songnian Lin for comments
on the preparation of the manuscript. We thank Drs. John Piwinski,
William Greenlee, and Catherine Strader for their strong support.
We thank Drs. Alexei Buevich and Li-Kang Zhang for technical
support.
References and notes
1. (a) Frantz, S. Nat. Rev. Drug Disc. 2006, 5, 977; (b) Krishan, M. Pharma Biotech
[Internet]. Knol. 2011, 125, 10.08.2011. Available from: http://knol.google.com/
k/m-a-review-pharmaceutical-biotechnology-industry.
2. Piachaud, B. S. Technovation 2002, 22, 81.
3. Melnikova, I.; Golden, J. Nat. Rev. Drug Disc. 2004, 3, 993.
4. Tobon, G. J.; Youinou, P.; Saraux, A. Autoimmun. Rev. 2010, 9, A288.
5. Arend, W. P.; Dayer, J. M. Arthritis Rheum. 1990, 33, 305.
6. Firestein, G. S.; Alvaro-Gracia, J. M.; Maki, R. J. Immunol. 1990, 144, 3347.
7. Rutgeerts, P.; D’Haens, G.; Targan, S.; Vasiliauskas, E.; Hanauer, S. B.; Present, D.
H.; Mayer, L.; Van Hogezand, R. A.; Braakman, T.; DeWoody, K. L.; Shaible, T. F.;
Van Deventer, S. J. H. Gastroenterology 1999, 117, 761.