Amide-based inhibitors of p38α MAP kinase. Part 2: Design, synthesis and SAR of potent N-pyrimidyl amides

Article abstract of DOI:10.1016/j.bmcl.2010.02.090

Optimization of a tri-substituted N-pyridyl amide led to the discovery of a new class of potent N-pyrimidyl amide based p38α MAP kinase inhibitors. Initial SAR studies led to the identification of 5-dihydrofuran as an optimal hydrophobic group. Additional

Full text of DOI:10.1016/j.bmcl.2010.02.090

Bioorganic & Medicinal Chemistry Letters 20 (2010) 2560–2563
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Amide-based inhibitors of p38a MAP kinase. Part 2: Design, synthesis and SAR
of potent N-pyrimidyl amides
*
Richland Tester, Xuefei Tan , Gregory R. Luedtke, Imad Nashashibi, Kurt Schinzel, Weiling Liang, Joon Jung,
*
Sundeep Dugar, Albert Liclican, Jocelyn Tabora, Daniel E. Levy , Steven Do
Department of Medicinal Chemistry, Scios Inc. 6500 Paseo Padre Parkway, Fremont, CA 94555, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Optimization of a tri-substituted N-pyridyl amide led to the discovery of a new class of potent N-pyrim-
Received 8 January 2010
Revised 23 February 2010
Accepted 24 February 2010
Available online 2 March 2010
idyl amide based p38a MAP kinase inhibitors. Initial SAR studies led to the identification of 5-dihydrofu-
ran as an optimal hydrophobic group. Additional side chain modifications resulted in the introduction of
hydrogen bond interactions. Through extensive SAR studies, analogs bearing free amino groups and alter-
natives to the parent (S)-
cellular activities and maintained balance between p38
a
-methyl benzyl moiety were identified. These compounds exhibited improved
and CYP3A4 inhibition.
Ó 2010 Elsevier Ltd. All rights reserved.
a
Keywords:
p38
MAP
Kinase
Antiinflammatory
p38 Mitogen-activated protein kinase (MAPK) is a serine–thre-
onine protein kinase and one of the best characterized enzymes in-
volved in the inflammatory process.1,2 Of the four isoforms (
a
, b,
is the most studied and perhaps the most physiolog-
ically relevant kinase involved in inflammatory responses.
p38 MAP kinase is activated in response to cellular stress,
growth factors and cytokines such as interleukin-1 (IL-1) and
tumor necrosis factor-alpha (TNF- ). Once activated, p38 stimu-
c,
and d), p38
a
a
Figure 1. Structure and potency of N-pyridyl amide lead 1.
a
a
lates cytokine production through activation of additional kinases
leading to phosphorylation of heat shock proteins and transcrip-
tion factors. Since modulation of cytokines has shown clinical effi-
cacy in treating inflammatory disorders such as rheumatoid
arthritis,3–5 inflammatory bowel disease,6 congestive heart failure7
and psoriasis,8 the ability of p38
a
to modulate cytokine production
has led to its implication in inflammation-based pathologies.2,9–13
Driven by the central role of p38 in settings of inflammation,
a
this enzyme has been an active target for drug discovery efforts
over the past decade.14–18 Contributing to these efforts, we
previously described initial studies focusing on a novel class of
N-pyridyl amides (Fig. 1).19
Figure 2. Structure and potency comparison between N-pyridyl amide 2 and N-
pyrimidyl amide 3.
Subsequent SAR studies involving incorporation of a (S)-a-
methyl benzyl moiety at the 2-position of the pyridine accompa-
nied by additional structural modifications led to the discovery
of compound 2.19 Later conversion of the pyridine to a pyrimidine
resulted in identification of compound 3 ( Fig. 2) possessing
significant improvements in potency against p38
matic and cell-based assays. Herein we describe our subsequent
SAR activities focused on the pyrimidine scaffold.
a in both enzy-
As illustrated in Figure 3, the proposed binding mode of the cis-
conformer of 3 in the ATP binding site of p38a MAP kinase involves
a hydrogen bond between a pyrimidine nitrogen and the amide NH
* Corresponding authors. Tel.: +1 011 86 22 66239656 (X.T.); tel.: +1 650 704
3051 (D.E.L.).
of Met-109. An additional hydrogen bond is noted between the
E-mail addresses: xuefei827@gmail.com (X. Tan), del345@gmail.com (D.E. Levy).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.02.090

R. Tester et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2560–2563
2561
Table 1
p38a enzyme and cellular activity for compounds 7a–v
Compd
R1
p38
a
IC50
(
l
M, n = 3)21
dWBA (
0.187
l
M, n = 3)22
3
0.064
0.017
0.164
0.071
7a
0.387
7b
7c
>1
0.313
7d
0.09
0.829
Figure 3. Proposed binding mode of 3 in the p38a
ATP binding site.20
7e
7f
0.045
0.180
0.3
>1
bridging NH and the Met-109 carbonyl. The naphthyl group occu-
pies the Thr-106 hydrophobic pocket known to be responsible for
p38a a-methyl benzyl moiety occupies
kinase specificity.23–25 The
7g
7h
0.188
0.154
>1
an additional hydrophobic pocket. The model also identifies a dis-
tal hydrogen bonding formed between the amide carbonyl of 3 and
the Lys-53 side chain. Based on this binding mode, we hypothesize
that potency enhancements may be achieved through incorpora-
tion of additional hydrogen bonds with neighboring amino acids,
such as Asp-112, Ser-154, Asp-168 and Tyr-35.
>10
7i
0.179
0.193
0.267
>10
>10
>1
The compounds of the present study were prepared as illustrated
in Scheme 1. As shown, reaction of 2,4-dichlo-pyrimidine with eth-
ylamine gave 2-chloro-4-ethylaminopyrimidine, 5. Deprotonation
with NaH followed by quenching with various acid chlorides al-
7j
7k
lowed development of the R1 SAR. Final treatment with (S)-
a-
7l
0.081
0.313
0.013
0.533
>10
methyl benzyl amine via palladium catalysis gave the desired final
products, compounds 7.26 The p38 activities for this series are sum-
marized in Table 1.
7m
7o
As shown in Table 1, compared to the 2-naphthyl analog, 3, com-
pounds bearing bicyclic groups such as 5-dihydrobenzofuryl, 7a,
and 5-indanyl, 7e, showed improved enzyme activity. In contrast,
analogs incorporating groups such as 3,4-(methylenedioxy)phenyl,
7b, 7-dihydrobenzofuryl, 7d, 3-benzo-thienyl, 7g, N-methyl-3-indo-
lyl, 7h, and 2-benzofuryl, 7i, were several fold less active. Regarding
compound 7f, the fourfold loss in potency compared to 7e may be
caused by steric effects due to a six-membered ring versus a five-
membered ring. Due to this hypothesis, a series of analogs bearing
monocyclic groups at R1 was prepared. This hypothesis was vali-
dated through the observation that substituted phenyl groups were
0.14
7p
7q
0.091
0.081
0.143
0.4
7r
7s
0.065
0.104
0.152
0.349
compatible with potent p38a inhibition. Of particular interest were
meta-substituted compounds 7l, 7o, 7r, and 7t—all showing activity
in the enzyme assay at approximately 80 nM or better. The two most
7t
0.064
0.039
>1
7u
0.178
7v
0.026
0.058
potent compounds (7u and 7v) were identified as the more metabol-
ically stable analogs of 7o. In reconciling the enzyme inhibitory
activity with the proposed binding mode in Figure 3, it appears that
the hydrophobic pocket formed by Thr-106, Val-105, and Leu-104
can tolerate both polar and non-polar substituents. However, when
polar groups reside in this pocket, activity is dependent upon the ori-
entation of the group.
Scheme 1. General synthesis of N-pyrimidyl amides 7. Reagents and conditions: (a)
ethyl amine, K2CO3, DMF, 45%; (b) NaH, DMF then acid chloride, 70–95%; (c) (S)-
benzyl amine, Pd(OAc)2, BINAP, Cs2CO3, dioxane, 110 °C, 30–80%.
a-

2562
R. Tester et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2560–2563
Also shown in Table 1, enzyme inhibitory activity did not neces-
sarily track with cell-based data. While our most potent analogs
bearing bicyclic groups at R1 inhibited at approximately
400–900 nM, our best monocyclic derivative, compound 7o, 7u,
and 7v inhibited at 140 nM, 178 nM, and 58 nM, respectively.
While these results showed promise, we felt that the benzofuran
group of 7a was able to optimally fill its respective hydrophobic
pocket. Consequently, this compound was chosen as a new base-
line for additional SAR studies targeting the amide side chain.
The compounds shown in Table 2 were prepared as illustrated
in Scheme 1 by substituting ethylamine with alternative amines.
In cases where final products contained free amino groups, these
amines were incorporated into Scheme 1 as Boc derivatives. Boc
cleavage was incorporated into the final step using standard TFA
conditions.26
As shown in Table 2, modifications at R2 generally resulted in
weaker inhibition compared to compound 7a. This observation
was independent of the presence of hydrogen bond donors or
acceptors. However, when R2 was modified to incorporate a chain
of four carbons, both enzymatic and cell-based potency improved.
In this series, the best analogs (compounds 8e and 8l) showed low
nM activity in both assays. As illustrated in Figure 4, the improved
enzymatic activity is explained through the introduction of hydro-
gen bonds with Ser-154 and the side chain of Asp-112.
While the SAR thus far resulted in improvements in both enzy-
matic and cellular inhibition, the broader set of compounds stud-
ied, of which only a subset is presented herein, were consistently
identified as potent inhibitors of CYP3A4. This CYP activity was
Figure 4. Proposed binding modes of 8e and 8l in the ATP binding site of p38
a
enzyme.20 For analog 8e, the methyl piperidine nitrogen is able to form a hydrogen
bond interaction with backbone carbonyl oxygen of Ser-154. For analog 8l, the butyl
amine nitrogen could potentially form two hydrogen bond interactions with Ser-
154 and the side chain of Asp-112.
traced to the (S)-a-methylbenzyl group. Thus, in order to proac-
tively design structures with minimal potential for drug–drug
interactions, structures bearing alternatives to this group were pre-
pared. These analogs, summarized in Table 3, were prepared
according to Scheme 1 by replacing (S)-a-methyl benzylamine
Table 3
Table 2
p38a activity and CYP3A4 profile for compounds 9a–10d
p38a enzyme and cellular activity for compounds 7a, 8a–n
Compd R3
p38
a
IC50
(lM,
dWBA (
lM,
CYP3A4 (lM,
n = 3)21
n = 3)22
n = 3)27
Compd R2
p38a
IC50
(l
M, n = 3)21 dWBA ( M, n = 3)22
l
7a
8a
0.017
0.462
0.387
2.088
8e
0.054
0.015
0.094
8b
8c
0.165
0.345
0.704
>10
9a
9b
0.024
0.36
0.088
3.29
2.9
>20
8d
8e
8f
0.158
0.054
>10
>10
0.015
—
9c
0.40
>10
>20
9d
9e
0.316
3.04
0.46
>10
2.02
1.54
8g
8h
8i
>10
—
9f
0.53
0.60
0.112
0.05
0.367
0.34
9g
0.711
0.673
0.524
>10
8l
0.004
0.009
0.18
10a
10b
0.017
0.011
0.09
3.115
>20
8k
8l
>10
—
0.865
0.004
0.096
0.009
0.13
8m
10c
10d
0.025
0.31
0.048
2.94
2.648
11.72
8n
0.072
0.564

R. Tester et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2560–2563
2563
with different amines.26 The parent structures for this SAR were
compounds 8e and 8l.
As shown in Table 3, replacing the (S)-a-methyl benzyl group
impacted not only CYP3A4 activity, but also enzymatic and cellular
15. Hynes, J., Jr.; Leftheris, K. Curr. Top. Med. Chem. 2005, 5, 967.
16. Goldstein, D. M.; Gabriel, T. Curr. Top. Med. Chem. 2005, 5, 1017.
17. Dominguez, C.; Powers, D. A.; Tamayo, N. Curr. Opin. Drug Discov. Devel. 2005, 4,
421.
18. Wagner, G.; Laufer, S. Med. Res. Rev. 2006, 26, 1.
19. Luedtke, G. R.; Schinzel, K.; Tan, X.; Tester, R. W.; Nashashibi, I.; Xu, Y.-j.; Dugar,
S.; Levy, D. E.; Jung, J. Bioorg. Med. Chem. Lett. doi:10.1016/j.bmcl.2010.02.088.
activity. Of the modifications shown, only a 2-propyl alternative
maintained p38a activity while reducing CYP3A4 activity by up
20. Docking studies were conducted using a high-resolution p38a X-ray structure
of p38 (PDB ID: 1ove). The extra-precision mode of Glide, a ligand-receptor
a
to two orders of magnitude compared to the enzymatic activity.
In the case of parent compound 8l, (S)-2-(1-methoxy)propyl was
also well tolerated.
In summary, high throughput screening identified N-pyridyl
amides as novel inhibitors of p38a. Through subsequent SAR stud-
ies, both enzymatic and cellular activities were optimized (com-
pounds 8e and 8l). Improved enzymatic activity is believed due
to optimal occupation of hydrophobic pockets and introduction
of additional hydrogen bonding interactions. Additional SAR pro-
duced structures with improved CYP profiles. These compounds
(analogs 9a, 10a, and 10c) were targeted as candidates for further
development.
docking tool from Schrodinger, LLC, was used to generate a number of poses for
analogs 3, 8e, and 8l, which were further filtered by visual inspection.
21. Mavunkel, B. J.; Perumattam, J. J.; Tan, X.; Luedtke, G. R.; Lu, Q.; Lim, D.; Kizer,
D.; Dugar, S.; Chakravarty, S.; Xu, Y.-j.; Jung, J.; Liclican, A.; Levy, D. E.; Tabora, J.
Bioorg. Med. Chem. Lett. 2010, 20, 1059.
22. The LPS/TNF
a Human Whole Blood Assay was run as described in Mavunkel,
B.; Dugar, S.; Luedtke, G.; Tan, X.; McEnroe, G. U.S. patent 6,696,443, 2004.
23. Davies, S. P.; Reddy, H.; Caivano, M.; Cohen, P. Biochem. J. 2000, 351, 95.
24. Wang, Z.; Canagarajah, B. J.; Boehm, J. C.; Kassisa, S.; Cobb, M. H.; Young, P. R.;
Abdel-Meguid, S.; Adams, J. L.; Goldsmith, E. J. Structure 1998, 6, 1117.
25. Lisnock, J.; Tebben, A.; Frantz, B.; O’Neill, E. A.; Croft, G.; O’Keefe, S. J.; Li, B.;
Hacker, C.; De Laszlo, S.; Smith, A.; Libby, B.; Liverton, N.; Hermes, J.; LoGrasso,
P. Biochemistry 1998, 37, 16573.
26. Tester, R.; Tan, X.; Schinzel, K.; Nashashibi, I.; Luedtke, G. R.; Liang, W.; Jung, J.;
Goehring, R. R.; Dugar, S.; Do, S. US 2006/199821, 2006.
27. CYP3A4 assay procedure: compounds were diluted from DMSO stocks to 40 lM
in cofactors/serial dilution buffer (0.05 M potassium phosphate, 2 mg/mL
NADP+, 2 mg/mL glucose-6-phosphate, 1.3 mg/mL magnesium chloride
hexahydrate, 0.8 U/mL glucose-6-phosphate dehydrogenase), for example,
References and notes
1. Han, J.; Lee, J.-D.; Bibbs, S. L.; Ulevitch, R. J. Science 1994, 265, 808.
2. Lee, J. C.; Kassis, S.; Kumar, S.; Badger, A.; Adams, J. L. Pharmacol. Ther. 1999, 82,
389.
3. Foster, M. L.; Halley, F.; Souness, J. E. Drug News Perspect. 2000, 13, 488.
4. Pargellis, C.; Regan, J. Curr. Opin. Invest. Drugs 2003, 4, 566.
5. Smith, R. J. Drug Discovery Today 2005, 10, 1598.
2
l
L 10 mM stock into 500
pipette 0.1 mL of cofactors/serial dilution buffer + 0.4% DMSO were dispensed
into the wells in columns 2–12. The diluted test compounds (150 L) or
positive control were added to the desired well in column 1. With
multichannel pipette, 0.05 mL from the wells in column was serially
lL = 40 lM in 0.4% DMSO. With a multichannel
l
a
1
diluted to the wells in columns 2 through 8. The solution was mixed by
pipetting 3–5 times in each well. The remaining 0.05 mL in the wells in column
8 was removed and discarded. A lid was placed on the plate and it was put at
37 °C in an incubator for at least 10 min. With a multichannel pipette, 0.1 mL
enzyme/substrate mix (100 pmol CYP3A4+OR+b5 Supersomes™, 50 mM 7-
benzyloxy-4-(trifluoromethyl)-coumarin, 0.5 M potassium phosphate) was
dispensed to columns 1 through 11. The liquid was dispensed in a stream to
facilitate mixing. The lid was replaced and incubation continued at 37 °C for
6. Badger, A. M.; Bradbeer, J. N.; Votta, B.; Lee, J. C.; Adams, J. L.; Griswold, D. E. J.
Pharmacol. Exp. Ther. 1996, 279, 1453.
7. Kubota, T.; Miyagishima, M.; Alvarez, R. J.; Kormos, R.; Rosenblum, W. D.;
Demetris, A. J.; Semigran, M. J.; Dec, G. W.; Holubkov, R.; McTiernan, C. F.;
Mann, D. L.; Feldman, A. M.; McNamara, D. M. J. Heart Lung Transplant. 2000, 19,
819.
8. Sticherling, M. Drug Discovery Today: Dis. Mech. 2005, 2, 275.
9. Herlaar, E.; Brown, Z. Mol. Med. Today 1999, 5, 439.
10. Lee, J. C.; Kumar, S.; Griswold, D. E.; Underwood, D. C.; Votta, B. J.; Adams, J. L.
Immunopharmacology 2000, 47, 185.
11. Kumar, S.; Boehm, J.; Lee, J. C. Nat. Rev. Drug Disc. 2003, 2, 717.
12. Saklatvala, J. Curr. Opin. Pharmacol. 2004, 4, 372–377.
13. Ashwell, J. D. Nat. Rev. Immunol. 2006, 33, 1447.
30 min. With
a miltichannel pipette, 0.075 mL of STOP solution (80%
acetonitrile/20% 0.5 M tris base) was dispensed to all wells. Dispense the
liquid in a stream to facilitate mixing. The enzyme/substrate mix (0.1 mL) was
dispensed to the wells in column 12 and the plate was scanned with a
fluorescent plate scanner (Perkin Elmer LS50B—excitation filter 409 nM,
emission filter 530 nM).
14. Diller, D. J.; Lin, T. H.; Metzger, A. Curr. Top. Med. Chem. 2005, 5, 953.