Benzothiophene inhibitors of MK2. Part 1: Structure-activity relationships, assessments of selectivity and cellular potency

Article abstract of DOI:10.1016/j.bmcl.2009.02.015

Identification of potent benzothiophene inhibitors of mitogen activated protein kinase-activated protein kinase 2 (MK2), structure-activity relationship (SAR) studies, selectivity assessments against CDK2, cellular potency and mechanism of action are pres

Full text of DOI:10.1016/j.bmcl.2009.02.015

Bioorganic & Medicinal Chemistry Letters 19 (2009) 4878–4881
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Benzothiophene inhibitors of MK2. Part 1: Structure–activity relationships,
assessments of selectivity and cellular potency
*
*
David R. Anderson , Marvin J. Meyers , Ravi G. Kurumbail, Nicole Caspers, Gennadiy I. Poda, Scott A. Long,
Betsy S. Pierce, Matthew W. Mahoney, Robert J. Mourey
Pfizer Global Research and Development, St. Louis Laboratories, 700 Chesterfield Parkway West, Chesterfield, MO 63017, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Identification of potent benzothiophene inhibitors of mitogen activated protein kinase-activated protein
kinase 2 (MK2), structure–activity relationship (SAR) studies, selectivity assessments against CDK2, cel-
lular potency and mechanism of action are presented. Crystallographic data provide a rationale for the
observed MK2 potency as well as selectivity over CDK2 for this class of inhibitors.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 6 October 2008
Revised 3 February 2009
Accepted 5 February 2009
Available online 8 February 2009
Keywords:
MK2
MAPKAP-K2
Mitogen activated protein kinase-activated protein kinase 2
(MK2) is a Ser/Thr protein kinase in the calcium/calmodulin-
dependent protein kinase family. MK2 is a direct substrate of p38
kinase and is important for cytokine production.1–3 MK2 knockout
5
X
6
O
7
R
3
NH
2
1
S
mice show reduced expression of TNFa when stimulated with lipo-
O
polysaccharide (LPS) and are resistant to developing disease in
1
arthritis models.2,3 MK2 knockout mice have a normal phenotype
Fig. 1. General structure of benzothiophene MK2 inhibitors.
suggesting that MK2 may be
intervention.
a safe target for therapeutic
A number of chemotypes have been reported as MK2 inhibitors
by us as well as others.4–9 In the current study we report a different
chemical class, benzothiophenes, with a general structure shown
in Figure 1.
This chemotype is similar to previously reported structures in
that it has a five-membered ring fused to a lactam.5–9 We antici-
pated a similar binding mode with the 7-methoxy group forming
a hydrogen bond in the hinge region (Leu141 backbone NH) and
the lactam portion binding to the conserved catalytic Lys93 and
Asp207 in the active site of MK2 (Fig. 2a).
The synthesis of this class is shown in Scheme 1. Cinnamic acid
2 was converted to intermediate 3 with thionyl chloride in chloro-
benzene.10 Direct displacement of the chlorides with amines was
not successful, but palladium catalyzed coupling of 3 and Boc-pro-
tected diamines 4–8 using ( )-BINAP as the catalyst gave the de-
sired adduct in good yields.11 Sequential deprotection with TFA
and cyclization with NaOMe/MeOH gave the target compounds.
Protected diamines 5–7 were prepared via borane reduction of
the primary amides of each enantiomer of alanine.12
Scheme 2 shows the synthesis of compounds designed to test
the role of the nitrogen atom at the 5-position and further elabora-
tion of compound 13. In order to replace the nitrogen with sulfur,
the chlorine on intermediate 3 was displaced with a protected ami-
no thiol 14 followed by deprotection and cyclization to yield ana-
log 15. For comparison, the carbon analog was prepared by treating
protected allyl amine 16 with 9-BBN followed by Suzuki coupling
to 3. Deprotection and cyclization of 3 produced analog 17. Alcohol
13 was activated as the mesylate and displaced with sodium azide.
Reduction with triphenyl phosphine gave the primary amine 18.
The potencies of the analogs described in Schemes 1 and 2 for
MK2, CDK2 and TNF
a production an LPS-stimulated U937 TNFa
cell assay are reported in Table 1.
Nitrogen containing analogs 9–11 were more potent than the
carbon and sulfur analogs (compare 9, 15, and 17). The configura-
tion of the stereogenic center in 10 and 11 appears to be important
with analog 10 being about 10-fold more potent against both MK2
and CDK2 compared to its enantiomer. Interestingly, this improve-
ment in enzyme potency is reflected in improved cell potency as
* Corresponding authors. Tel.: +1 636 247 7651 (D.R.A.).
E-mail address: david.r.anderson@pfizer.com (D.R. Anderson).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.02.015

D. R. Anderson et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4878–4881
4879
well. Compounds 13 and 18 introduce a heteroatom onto the diaz-
apenone substituent resulting in a gain in potency for MK2 and in
the case of 18 an improvement in CDK2 selectivity. However, only
for alcohol 13 is this improvement in enzyme potency manifested
in an improvement in cell potency. Presumably this is due to re-
duced permeability of the analog with the primary amine being
predominantly ionized in solution. Extension of the chain without
a heteroatom did result in a loss of potency (compound 12) sug-
gesting that the heteroatom (oxygen or nitrogen) could be engaged
in a direct or water-mediated hydrogen bonding interaction with
the protein.
Most of these compounds (with the exception of 18) are not
selective for MK2 over CDK2: they are nearly equipotent to both
MK2 and CDK2. Broad kinase panel profiling of 10 revealed that
it is a potent inhibitor of numerous kinases with less than 10-fold
selectivity against 18 of the 90 kinases profiled. To better under-
stand contributions to potency and selectivity, X-ray crystal struc-
tures of compound 18 were obtained with MK2 (at 3.5 Å
resolution) and CDK2 (at 1.9 Å resolution) co-crystals (Fig. 2).
While the resolution of the MK2 structure is not very high, some
analysis can be done to account for the discrepancy of the observed
potencies.
Although MK2 and CDK2 are quite similar in the phosphate re-
gion (with the exception of Thr206 in MK2 vs Ala144 in CDK2) the
aminomethyl does not bind in a similar fashion to both kinases. In
the MK2 structure, it forms a hydrogen bond to the backbone car-
bonyl of Glu190 in the solvent exposed region while in the CDK2
structure it forms a hydrogen bond with both side chains of the
catalytic Asp145 and conserved Asn132. In the MK2 structure the
lactam nitrogen is hydrogen bonded to the catalytic Asp207 while
the lactam carbonyl is engaging catalytic Lys93. It is worth empha-
sizing that the lactam nitrogen is not engaged in a hydrogen bond
interaction with the catalytic Asp145 in the CDK2 structure since
the Asp145 side chain has undergone a conformational change
and is interacting with the aminomethyl of compound 18 instead.
However, in both structures the lactam carbonyl interacts with the
catalytic Lys while the 7-methoxy group forms a hydrogen bond to
the hinge region. The hydrogen bond to the hinge, which is also
considered as a key interaction in a large number of kinases, is a
bit longer (3.3 Å) in the CDK2 structure compared to that of MK2
Fig. 2. X-ray crystal structures of 18 in the complex with a) MK2 (3.5 Å resolution,
3FYK) and b) CDK2 (1.9 Å resolution, 3FZ1). The structures revealed different
engagement of the aminomethyl in the two protein kinases that provides rational
for selectivity for MK2 over CDK2.
MeO
NHBoc
MeO
Cl
MeO
Cl
MeO
CO2H
a
16
OMe
OMe
S
S
b
NH
O
O
3
S
2
3
O
O
NHBoc
R1 R2
NHBoc
NHBoc
a
H2N
O
17
HS
N Boc
HO
R1 R2
14
b
4: R1 = H; R2 = H
5: R1 = Me; R2 = H
6: R1 = H; R2 = Me
7: R1 = n-Pr; R2 = H
8
MeO
H2N
R1
R2
NH
c
S
NHBoc
R1 R2
NH
HN
S
H2N
S
15
MeO
O
5: R1 = Me; R2 = H
6: R1 = H; R2 = Me
7: R1 = n-Pr; R2 = H
O
9: R1 = H; R2 = H
MeO
HN
10: R1 = Me; R2 = H
11: R1 = H; R2 = Me
c
NH2
13
NH
12: R1 = n-Pr; R2 = H
S
13: R1 = CH2OH; R2 = H
O
18
Scheme 1. Reagents and conditions: (a) (i) SOCl2, chlorobenzene, 120 °C; (ii)
MeOH; (b) (i) Pd2(dba)3 (5 mol %), ( )BINAP (10 mol %), Cs2CO3 (2 equiv) toluene,
110 °C, 24 h; (ii) TFA/CH2Cl2; (iii) NaOMe/MeOH; (c) (i) CDI; (ii) NH4OH; (iii)
BH3ÁTHF.
Scheme 2. Reagents and conditions: (a) (i) DBU/DMF; (ii) TFA/CH2Cl2; (iii) NaOMe/
MeOH; (b) (i) 9-BBN, 12, THF 0 °C; (ii) Pd(PPh3)4, Na2CO3; (iii) TFA/CH2Cl2; (iv)
NaOMe/MeOH; (c) (i) MsCl, NEt3, CH2Cl2; (ii) NaN3, DMF; (iii) PPh3, THF, H2O.

4880
D. R. Anderson et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4878–4881
Table 1
MeO
HN
MK2 and CDK2 potencies of analogs 9–13, 15, 17, 18
a
R
13
5
X
R1
NH
6
O
S
3
7
R2
NH
O
2
R =
N
H
19
N
H
N
H
1
S
O
20
21
Compound
number
X
R1, R2
MK2
inhibition
CDK2
inhibition IC50
M)
U937 TNFa
a
a
release IC50
(lM)
N
a
N
IC50
(l
M)
(l
N
O
9
N
N
N
N
N
S
H,H
0.18
0.04
0.30
1.52
0.16
0.012
0.26
1.4
0.7
3.8
5.45
0.4
5.5
4.7
2.6
10
11
12
13
15
17
18
Me, H
H, Me
n-Pr, H
22
23
24
—
Scheme 3. Reagents and conditions: (a) (i) MsCl, NEt3, CH2Cl2; (ii) amines, K2CO3,
KI, 100 °C, microwave.
CH2OH, H 0.014
0.025
0.15
0.36
H,H
H,H
1.09
0.50
C
N
CH2NH2,H 0.005
0.146
Table 2
a
Values are means of at least three experiments and standard deviations were
within 50% of the reported value. Assays conditions are the same as described in
Ref. 5.
MK2 and CDK2 potencies of analogs 19–24
Compound
number
MK2 inhibition
CDK2 inhibition
U937 TNFa release
IC50 (lM)
a
a
a
IC50
(
l
M)
IC50
(lM)
19
20
21
22
23
24
0.57
0.45
0.57
>20
>20
1.91
3.1
2.6
2.7
—
—
3.17
21.6
14.9
21
—
—
(2.8 Å) which may account to a certain extent for the discrepancy
in potency. While the nitrogen at the 5-position seems to be
important for potency against both MK2 and CDK2, it does not ap-
pear to be making a direct hydrogen bond to either kinase. How-
ever, it might form a water-mediated hydrogen bond to the
backbone carbonyl of Ile10 of MK2 and of Leu70 of CDK2. Another
contributing factor could be the 5-position nitrogen donates some
electron density to the lactam carbonyl oxygen via the thiophene
ring. Alternatively, it could be also influencing the conformation
of the seven-membered lactam ring (sp2 nitrogen vs sp3 carbon)
to facilitate hydrogen bonding interaction of the lactam carbonyl
and NH to the MK2 catalytic Asp/Lys pair.
In an attempt to improve the cell potency of 18, additional ami-
no substituents were prepared and evaluated as described in
Scheme 3. Alcohol 13 was activated as the mesylate and displaced
with amines. The results are summarized in Table 2.
Simple alkyl substitution resulted in about 100-fold loss in po-
tency for MK2 and a loss of cell potency. Cyclic tertiary amines 22
and 23 had no measurable potency towards MK2. None of the
tested analogs had superior potency compared to the unsubstitut-
ed diazapenone 9.
24
a
IC50 values are means of at least three experiments and standard deviations
were within 50% of the reported values. Assays conditions are the same as described
in Ref. 5.
R
O
HO
HN
HN
a
b
10
NH
NH
S
S
O
O
25
26-31
O
R =
O
26
27
28
In our previous report5 we described a strategy for gaining
selectivity against CDK2 by exploiting differences in the amino acid
sequence in the hinge region between MK2 and CDK2. Alteration of
the hinge binding element was also examined in an effort to im-
prove MK2 potency, selectivity and cellular potency. The 7-meth-
oxy group of 10 was de-methylated to yield phenol 24. Further
elaboration gave analogs 26–31. (Scheme 4, potency data is in Ta-
ble 3).
O
O
30
31
29
Scheme 4. Reagents and conditions: (a) (i) MsCl/NEt3; (ii) amines
Table 3
MK2 and CDK2 enzyme potencies of analogs 25–31
None of these modifications afforded a compound with im-
proved selectivity, enzyme or cellular potency as compared to
compound 10. Only compound 29 was equally potent against
MK2, but with no improvements in selectivity against CDK2. The
lack of selectivity of these analogs may be due to the conforma-
tional flexibility of the ether groups. A detailed study of more rigid
systems is the subject of Part 2 of this Letter.
Compound
number
MK2 inhibition
CDK2 inhibition
U937 TNFa release
IC50 (lM)
a
a
a
IC50
(l
M)
IC50
(lM)
25
26
27
28
29
30
0.19
0.11
1.32
0.27
0.03
3.44
0.026
0.064
0.052
0.081
0.011
3.09
3.5
4.6
27
3.1
1.8
78
To verify that suppression of TNFa production in the U937 cell
a
assay correlated with MK2 inhibition, we determined the amount
of phospho-Ser78 present on HSP27 in LPS-stimulated U937 cells
pre-incubated with 13. HSP27 is a substrate of MK2 and may be
IC50 values are means of at least three experiments and standard deviations
were within 50% of the reported values. Assays conditions are the same as described
in Ref. 5.
used as an intracellular target biomarker to correlate with TNF
a
production as previously shown.5 Amounts of phospho-p38 and
phospho-JNK2 were also determined. Compound 13 showed no ef-
fect on levels of phospho-p38 and phospho-JNK2, but did inhibit
the phosphorylation of HSP27 in a dose- dependant manner with
an IC50 value within twofold of the TNFa IC50 value. Since 13 inhib-
ited the downstream marker (HSP27) but not the upstream kinase

D. R. Anderson et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4878–4881
4881
2. Winzen, R.; Kracht, M.; Ritter, B.; Wilhelm, A.; Chen, C.-Y. A.; Shyu, A.-B.;
Muller, M.; Gaestel, M.; Resch, K.; Holtmann, H. EMBO J. 1999, 18, 4969.
3. Gaestel, M. Nat. Rev. Mol. Cell Biol. 2006, 7, 120.
4. Anderson, D. R.; Hegde, S.; Reinhard, E.; Gomez, L.; Vernier, W. F.; Lee, L.; Liu, S.;
Sambandam, A.; Snider, P. A.; Masih, L. Bioorg. Med. Chem. Lett. 2005, 15, 1587.
5. Anderson, D. R.; Meyers, M. J.; Vernier, W. F.; Mahoney, M. W.; Kurumbail, R.
G.; Caspers, N.; Poda, G. I.; Schindler, J. F.; Reitz, D. B.; Mourey, R. J. J. Med. Chem.
2007, 50, 2647.
6. Wu, J.-P.; Wang, J.; Abeywardane, A.; Andersen, D.; Emmanuel, M.; Gautschi, E.;
Goldberg, D. R.; Kashem, M. A.; Lukas, S.; Mao, W.; Martin, L.; Morwick, T.;
Moss, N.; Pargelis, C.; Patel, U. R.; Patnaude, L.; Peet, G. W.; Skow, D.; Snow, R. J.;
Ward, Y.; Werneburg, B.; White, A. Bioorg. Med. Chem. Lett. 2007, 17, 4664–
4669.
7. Trujillo, J. I.; Meyers, M. J.; Anderson, D. R.; Hegde, S.; Mahoney, M. W.; Vernier,
W. F.; Buchler, I. P.; Wu, K. K.; Yang, S.; Hartmann, S. J.; Reitz, D. B. Bioorg. Med.
Chem. Lett. 2007, 17, 4657.
8. Goldberg, D. R.; Choi, Y.; Cogan, D.; Corson, M.; DeLeon, R.; Gao, A.; Gruenbaum,
L.; Hao, M. H.; Joseph, D.; Kashem, M. A.; Miller, C.; Moss, N.; Netherton, M. R.;
Pargellis, C. P.; Pelletier, J.; Sellati, R.; Skow, D.; Torcelini, C.; Tseng, Y.-C.; Wang,
J.; Watsi, R.; Werneburg, B.; Wu, J. P.; Xiong, Z. Bioorg. Med. Chem. Lett. 2008, 18,
938.
9. Xiong, Z.; Gao, D. A.; Cogan, D. A.; Goldberg, M.-H. H.; Moss, N.; Pack, E.;
Pargelis, C.; Skow, D.; Trieselmann, T.; Werneburg, B.; White, A. Bioorg. Med.
Chem. Lett. 2008, 18, 1994.
pathway (p38) it is reasonable to conclude that the observed cell
potency was primarily due to inhibition of MK2, thus, confirming
that the inhibition of TNFa production with our MK2 inhibitors is
within the mechanism of action.
In summary, we have reported an additional chemical series of
potent MK2 inhibitors with improved cellular potency and that the
observed cellular potency is within the mechanism of action.
Improvements in MK2 potency may be achieved by modification
of the diazapenone substituents. However, these changes did not
result in improved cellular potency. Examination of crystal struc-
tures revealed additional opportunities for improvement of selec-
tivity for MK2 over CDK2. It was observed that the engagement
of the protonated aminomethyl is slightly different in MK2 com-
pared to CDK2 resulting to a weaker hydrogen bond between the
7-methoxy group to the hinge for CDK2 compared to MK2 (0.5 Å
difference in the length of the hydrogen bond). Discovery of MK2
selective inhibitors within this series will be the focus of Part 2.
References and notes
10. Higa, T.; Krubsack, A. J. Org. Chem. 1976, 41, 3399.
11. Ferreira, I. C. F. R.; Queiroz, M.-J. R. P.; Kirsch, G. Tetrahedron 2003, 59, 975.
12. Cho, H.; Gogliotti, R. D.; Hamilton, H. W.; Krasutsky, A.; Nakamura, T.; Tade, H.;
Weber, P., C. WO 2002014324, 2002.
1. Kotlyarov, A.; Neininger, A.; Schubert, C.; Eckert, R.; Birchmeier, C.; Volk, H.-D.;
Gaestel, M. Nat. Cell Biol. 1999, 2, 94.