Optimization of α-ketoamide based p38 inhibitors through modifications to the region that binds to the allosteric site

Article abstract of DOI:10.1016/j.bmcl.2010.06.102

We have optimized a novel series of potent p38 MAP kinase inhibitors based on an α-ketoamide scaffold through structure based design that due to their extended molecular architecture bind, in addition to the ATP site, to an allosteric pocket. In vitro ADM

Full text of DOI:10.1016/j.bmcl.2010.06.102

Bioorganic & Medicinal Chemistry Letters 20 (2010) 4819–4824
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Optimization of
a
-ketoamide based p38 inhibitors through modifications to
the region that binds to the allosteric site
*
Antonio Garrido Montalban , Erik Boman, Chau-Dung Chang, Susana Conde Ceide, Russell Dahl,
David Dalesandro, Nancy G. J. Delaet, Eric Erb, Justin T. Ernst, Andrew Gibbs, Jeffrey Kahl, Linda Kessler,
Jeff Kucharski, Christopher Lum, Jan Lundström, Stephen Miller, Hiroshi Nakanishi, Edward Roberts,
Eddine Saiah, Robert Sullivan, Jan Urban, Zhijun Wang, Christopher J. Larson
Drug Discovery, Kémia, Inc., 5871 Oberlin Drive, Suite 100, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We have optimized a novel series of potent p38 MAP kinase inhibitors based on an a-ketoamide scaffold
Received 3 March 2010
Revised 14 June 2010
Accepted 21 June 2010
Available online 25 June 2010
through structure based design that due to their extended molecular architecture bind, in addition to the
ATP site, to an allosteric pocket. In vitro ADME, in vivo PK and efficacy studies show these compounds to
have drug-like characteristics and have resulted in the nomination of a development candidate which is
currently in phase II clinical trials for the oral treatment of inflammatory conditions.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
p38 MAP kinase
Antiinflammatory
SAR
Ketoamides
Cytokines
Drug-like
After the ground breaking work by SmithKline Beecham on
ATP-competitive or orthosteric (Type I) p38 inhibitors that led to
SB203580,1 and on allosteric (Type II) p38 inhibitors by BI that
led to BIRB-796,2 considerable research has been devoted to the
identification of additional inhibitors of these two distinct classes.3
This notion stems from the fact that p38 has been recognized as a
highly attractive target for therapeutic intervention. It is well
established that the proinflammatory cytokines tumor necrosis
N
O
O
N
O
H
N
O
S
N
H
N
N
H
N
N
BIRB-796
SB203580
F
factor-a (TNF-a) and interleukin-1b (IL-1b) play an important role
in the pathogenesis of various inflammatory diseases and that the
stress-activated signal transduction pathway leading to these cyto-
kines is in part regulated by p38. Blockade of TNF-a by the biolo-
Type II inhibitors interact, in addition to the active site, with a
region on the kinase that is spatially distinct from the ATP pocket.
Furthermore, they inhibit p38 MAP kinase via a conformation
(DFG-out) that is incompatible with ATP binding. As the DFG-in,
active conformational states of all kinases must be competent to
bind ATP, pharmacophores optimized to bind this structural motif
in one kinase may have some cross-reactivity with this structural
motif in other kinases, since both binding pockets must bind ATP
sufficiently to support catalysis. However, there is no comparable
selective pressure on the allosteric binding pocket of any two
kinases adopting a DFG-out conformation to bind similar pharma-
cophores. Thus, binding to and stabilization of the DFG-out, inac-
tive conformation should, in principle, lead to improve selectivity
across the kinome.8 In addition, selectivity for ATP site binders
gics EnbrelÒ, RemicadeÒ, and HumiraÒ in the treatment of
deseases such as rheumatoid arthritis, Crohn’s disease and psoria-
sis has led to clinical and commercial success.4 Consequently, the
pharmaceutical5 and medicinal chemistry6 communities anticipate
that p38 inhibitors will display a similar therapeutic benefit but
with the convenience of oral administration. Despite the advance-
ment of a number of p38 inhibitors into the clinic, however, a small
molecule therapeutic utilizing this mechanism of action still re-
mains elusive.7
* Corresponding author. Tel.: +1 619 519 1787.
E-mail address: catalaba@hotmail.com (A.G. Montalban).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.06.102

4820
A. G. Montalban et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4819–4824
Table 1
under physiological conditions may be quite different from what is
observed in vitro due to ATP having different affinities for kinases
and being present at high concentrations.9
P38a
inhibition data for 6-membered-ring
a-ketoamide derivatives
R1
The above considerations led us to believe that it could be an
attractive strategy to deliver a small molecule p38 therapeutic. We
recently reported a new structural class of potent, Type II, p38 inhib-
2
2
3
1
6
O
3
4
2
3
O
N
O
6
R2
1
N
H
X
itors based on an a
-ketoamide scaffold (1 X = O)10 related and that
8
4
5
5
6
R3
5
bind in a similar manner to BIRB-796. Furthermore, we showed that
reversing the ketoamide order as in 2 (Z = O), adds benefits such as
chemical flexibilityand improvedpotencies while retaining the allo-
steric binding mode.11 Herein, we now report that maintaining the
pharmacophore group as in 1 but embedding the moiety that occu-
pies the exposed Phe169 pocket into a six-membered-ring increases
activity against p38. In addition, we show that, as with our second
generation series 2, potency is not compromised when replacing
the ethoxy- for more rigid-spacers/pharmacophores. The overall
favorable drug-like properties of this new family of potent allosteric
7
a
Compound
R1
R2
R3
X
P38
0.25
0.033
0.028
0.042
0.25
0.070
0.059b
0.071c
0.20
a
IC50
(l
M)
3
4
5
6
7
8
9
10
11
12
13
t-Bu
t-Bu
H
H
H
H
CO
CO
CO
CO
CO
CO
CO
CO
CN
CN
CN
H
Morpholinyl
t-Bu
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
Cl
NHSO2R4
CONHR4
a-ketoamidep38inhibitorsultimatelyled totheselectionofa devel-
NHSO2Me
NHSO2Me
NHSO2Me
CNOH
CHOH
CH2
opment candidate which is currently in phase II clinical trials for the
treatment of various autoimmunedeseases. During the course of our
studies, BI published a similar approach and SAR studies around the
BIRB-796 class of compounds.2b
0.39
0.19
a
IC50 for inhibition of p38 kinase enzymatic activity. Data were generated either
by ELISA using commercially available human p38 and myelin basic protein as a
substrate, or using an Invitrogen Z’lyte kit with its proprietary substrate. IC50’s given
represent the means of a minimum of two, and generally three or more, indepen-
dent experiments. Standard deviation for assays typically 30% of the mean or less.
R2
R4 = Me.
b
O
O
O
N
R4 = cyclopropyl.
c
N
N
H
O
X
N
H
N
R
Y
X
R1
Z
the right size for the Phe169 hydrophobic pocket. With small alkyl
groups (e.g., Me, CF3) or larger alkyl groups (e.g., norbornyl, CMe2R
2
1
(R = Et, Pr or Ph), cyclohexyl) inhibition of TNF
generally decreased by more than 30%. Since efficacy and IC50’s of
compounds tested in both, the cellular TNF and P38 inhibitory
assays showed good correlation, we used efficacy in the cellular
TNF inhibitory assay as a first pass for our compound progression.
In addition, the t-butyl group seems to be optimally placed within
the six-membered-ring, since analogues with the t-butyl group in
the 2- or 4-instead of the 3-position of the aniline, as defined here,
were substantially less efficacious (again, generally by more than
a production was
Ketoamides 3–10 were synthesized, as shown in Scheme 1,
through reaction of the corresponding aniline derivatives 14 with
the acid chloride 15,10 whereas 13 was prepared from acid chloride
16.10 Oxime 11 (X = C@NOH) was readily prepared as a mixture of
geometrical isomers by treating 9 with hydroxylamine in the pres-
ence of pyridine. Reduction of 9, on the other hand, gave the alco-
hol 12 (X = CHOH).
a
a
a
As expected from the SAR trends we established previously for
our first and second generation a
-ketoamide p38 inhibitors,10,11 an
30% less inhibition of TNFa production). Docking studies of 3 also
unsubstituted phenyl- (Ar = Ph), saturated six-membered-ring or
revealed that when binding allosterically to p38, the 5-position
of the aniline ring is in close proximity to an arginine-rich region
of the protein. We reasoned that placing a hydrogen bond donor
and/or acceptor moiety near that region could result in additional
binding interactions. Indeed, the introduction of a cyano group into
the 5-position of the six-membered ring (4), for example, resulted
in a 7–8-fold improvement in potency over the parent compound
3. When the t-butyl group was replaced for an unsaturated hetero-
cyclic five- or six-membered-ring potency was retained as exem-
plified in compound 5. A clear advantage of a phenyl versus an
amino substituted five-membered heteroaromatic-ring as in 1
and 2 is chemical variability and accessibility. Replacement of
the t-butyl group as above, for example, proved synthetically much
more challenging with our first and second generation series as did
taking advantage of the binding interaction with the arginine-rich
region of the protein. By analogy with BIRB-7962 and as our mod-
eling suggested, the group at position one (1, R) of the pyrazole
ring was hypothesized to be in close proximity and could, there-
fore, favorably interact with the hydrophobic portion of the side
chain of the conserved residue Glu71.10 When we introduced a
methoxy group into the 6-position of the aniline, however, no fur-
ther improvement in potency was obtained (6) although it was ex-
pected to project into the same region of the protein, namely the
conserved residue of Glu71, according to our docking studies.
The compound lacking the cyano group (7), on the other hand,
had a similar potency to the parent compound 3, indicating that
simple alkyl groups in place of Ar resulted in no significant inhibi-
tion of p38a (data not shown). Introducing the t-butyl group into
the 3-position of the aniline, however, produced a compound (3,
Table 1) within the same potency range of our first generation
p38 inhibitors (e.g., 1 R = p-tolyl, X = O, p38
a IC50 = 0.32 l
M).10
Molecular modelling12 suggests that, similarly to BIRB-796,2 the
t-butyl group in 3 occupies the exposed Phe169 hydrophobic pock-
et and is essential for allosteric binding. Consistent with our first
generation series,10 modifications of the t-butyl group resulted in
less p38
a
inhibitory activity confirming again that it has about
O
N
O
Ar
O
NH2
O
N
14
Ar
iii
O
N
H
X
i
O
X
Cl
3-10 X = CO
ii
11 X = CNOH
12 X = CHOH
13 X = CH2
15 X = CO
16 X = CH2
Scheme 1. Reagents and conditions: (i) EtOAc, 0.5 N NaHCO3, 60 °C, 16 h; (ii)
NH2OHÁHCl, EtOH, pyridine, 45 °C, 12 h; (iii) MeOH, NaBH4, rt, 1 h.

A. G. Montalban et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4819–4824
4821
a substituent at the 6-position is at least tolerated. Small hydro-
phobic groups in conjunction with the methoxy substituent such
as in 8 confirmed the hypothesis of a small group at the 6-position
of the aniline been tolerated. Nevertheless, substitution at the 6-
position while maintaining the hydrogen bond donor and/or
acceptor group at the 5-position proved very useful for the adjust-
ment of drug-like properties, such as for example improved aque-
ous solubility, other than potency. Only small groups, however,
about the size of a methoxy group were generally preferred. For
example, substituting the methoxy group for OEt, OiPr or OiBu
replacement of the naphthyl- for a phenyl-ring (with or without
substitution, e.g., plain phenyl analog of
(R4 = Me), p38
IC50 = 7.3 M), alkylation of the amide nitrogen (inhibition of
TNF- production generally dropped below 10%), removal (e.g.,
simple OMe derivative of 9, TNF- IC50 = 3.1 M) or changing the
position of the ethoxymorpholino group (e.g., moving the ketoma-
mide moiety to the 2-position of the naphthyl-ring with respect to
the ethoxymorpholino group would generally decrease inhibition
9
a
l
a
a
l
of TNF-
(generally, >50-fold decrease in potency with respect to p38
substitution at the 3- and/or 5-positions (generally, 2–10-fold less
potent on p38 ) and replacement of the morpholino group for other
heterocyclic- (generally, 6–100-fold less potent on p38 ) and ring-
open-analogues (generally, 5–10-fold less potent on p38 ). In
a
production to ꢀ10%), homologation of the ethyl linker
a),
generally decreased the inhibition of TNFa production by 30–
50%. Other combinations of hydrogen bond donor/acceptor groups
at the 5-position in conjunction with the methoxy substituent at
a
a
the 6-position that retained high potency against p38
a
were, for
a
example, the sulfonamide and amide derivatives 9 (R4 = Me) and
10 (R4 = cyclopropyl), respectively. With both pharmacophore moi-
eties, only small hydrophobic groups (e.g., methyl, ethyl or cyclo-
propyl) at the R4 position were accepted into this pocket of the
protein. Reversing the sulfonamide or amide order at the 5-posi-
tion of the aniline of 9 and 10 did not compromise potency greatly.
Substitution of the 5-position with –SO2NMe2 or –NHCOMe, for
example, resulted in compounds with similar potencies to 9 and
10, respectively. Interestingly, oximes such as 11 generally trended
summary, these results are consistent with an allosteric mode of
binding similar to BIRB-7962 in which the t-butyl group occupies
the exposed Phe169 hydrophobic pocket and the amide simulta-
neously hydrogen bonds via the carbonyl with the backbone amide
hydrogen of Asp168 and through the N–H with the carboxylate of
Glu71. The second carbonyl group, on the other hand, acts as a
hydrogen bond acceptor for the amide hydrogen of the conserved
Lys53 residue which would otherwise form a salt-bridge with the
carboxylate of Glu71. The substituted ring of the naphthyl main-
tains the ideal edge-to-face interaction with Phe169 while the sec-
ond one resides deep within the kinase specificity pocket. The
morpholino group, on the other hand, forms a conserved hydrogen
bond with the backbone amide hydrogen of Met109. In addition,
the substituent at position 5 of the aniline ring is hypothesized
to interact with an arginine-rich region of the protein and the
group at the 6-position with the hydrophobic portion of the side
chain of the conserved residue of Glu71. Figure 1 shows the docked
structure of compound 10 superimposed over BIRB-796.13
Consistent with our second generation ketoamide p38 inhibi-
tors,11 more rigid-spacers/pharmacophores within the ATP binding
site (Table 2) gave sub 100 nM compounds comparable to the sub-
series having the ethoxy linker (see Table 1). Again, we established
that the nitrogen at the 4-position of the pyrimidine ring, as defined
here, is responsible for the hydrogen bond interaction with the back-
bone amide hydrogen of Met109 (17). This was further confirmed by
the fact that the pyridine analogues 18 (Y = CH) and 19 (Y = CH) were
as potent, whereas the pyrimidine regioisomer with the nitrogen at
the 6- instead of the 4-position was significantly less active against
to decreased p38
a inhibition when compared to their respective
keto-analogues. In contrast, the equivalent change in the first
and second generation series consistently gave 5–10-fold more po-
tent compounds (e.g., 1 R = p-tolyl, X = NOH, p38
and 2 XY = N, R1 = m-tolyl, Z = NOH, R2 = ethoxy morpholino, p38
IC50 = 0.059
M).10,11 This improvement in potency was postulated
a IC50 = 0.023 lM
a
l
to be due to the hydrogen bond interaction observed in the docked
structures of the oximes O–H with one of the amide carbonyl
groups of the conserve residue of Glu71. On the other hand, com-
putational models suggests that the extended (as the example in
Fig. 1 illustrates) molecular architecture in which 11 is locked does
not allow for the sulfonamide at position five of the aniline and the
oxime O–H to simultaneously optimally interact with their respec-
tive regions of the protein. Reduction to the alcohol (12) or removal
of the carbonyl group (13) resulted in reduced potencies and was
again consistent with the results we obtained previously with
our first and second generation series.10,11 Other changes that re-
sulted in decreased potencies and followed the same SAR trends
as our first and second generation series10,11 were substitution at
the 6- and/or 7-positions (e.g., analog of 9 (R4 = Me) with Cl in
p38a (this change generally resulted in compounds which were
the 6-position of the naphthyl-ring, TNF-a IC50 = 0.22 lM) or
Table 2
P38 inhibition data for six-membered-ring
atom linked rigid-spacers/phamacophores in the ATP binding site
a
a-ketoamide derivatives with hetero-
X
Y
5
R2
3
1
O
6
N
R1
N
H
O
OMe
a
Compound
R1
X
Y
R2
P38aIC50 (lM)
17
18
19
20
21
NHSO2Me
NHSO2Me
CN
NHSO2Me
NHSO2Me
NH
NH
NH
NH
O
N
H
H
H
0.072
0.025
0.051
0.008b
0.038
CH
CH
N
NHMe
morpholinyl
N
a
IC50 for inhibition of p38 kinase enzymatic activity. Data were generated either
by ELISA using commercially available human p38 and myelin basic protein as a
substrate, or using an Invitrogen Z’lyte kit with its proprietary substrate. IC50’s given
represent the means of a minimum of two, and generally three or more, indepen-
Figure 1. Docked structure of compound 10 in thick tube drawing to p38
a using
dent experiments. Standard deviation for assays typically 30% of the mean or less.
the induced fit docking protocol based on the X-ray structure of p38 cocrystalled
a
b
IC50 of LPS-stimulated TNF-
monocytic lineage (THP-1).
a production in immortalized human cells of a
with BIRB-796 (pdb1kv2.ent). Only residues within 5 Å from compound 10 are
shown for clarity. BIRB-796 is shown in magenta.

4822
A. G. Montalban et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4819–4824
Table 3
ꢀ10-fold less potent). Molecular modeling was also in agreement
P38a inhibition data for six-membered-ring a-ketoamide derivatives with directly
with this hypothesis. Small substituents at the 3-position of the
pyrimidine- or pyridine-ring, which for synthetic reasons limited
us to primary-, secondary- and tertiary-amines, were generally well
tolerated (20) with no difference between the heteroatom linking
the naphthyl- to the six-membered-heteroaromatic-ring (21).
Compounds in which the naphthyl- was directly attached to the
six-membered-heteroaromatic-ring were also similarly potent
(Table 3). The nitrogen was tolerated in both, the 3(X)- and 4(Y)-posi-
tions of the pyridine ring as exemplified with compounds 22 and 23,
respectively. Not surprisingly, decreasing the electron density of the
pyridine nitrogen responsible for the hydrogen bond interaction
with Me109, such as in 24, trended towards decreased potency
whereas activity was retained with electron donating groups such as
in 25. Again, small auxophore groups were preferred. Interestingly,
replacing the nitrile group of 22 with a sulfonamide resulted in a
5–10-fold decrease in potency. Substituting the 4-position of the
pyridine ring with a hydrogen bond acceptor moiety (see Fig. 2,
e.g., n = 0–1, NR1R2 = morpholinyl, NH-pyranyl, NHMe), however,
restored the sub 100 nM potency. As was the case with the oxime
11 (vide supra), molecular modeling suggests that the sulfonamide
and the nitrogen at the 3-position of the pyridine ring are not in a
suitablegeometricarrangementwithrespecttooneanothertoallow
them to simultaneously interact favorably with their respective re-
linked rigid-spacers/pharmacophores in the ATP binding site
X
Y
O
NC
N
H
OMe
O
a
Compound
X
Y
P38aIC50 (lM)
22
23
24
25
N
CH
N
CH
N
CF
N
0.078
0.060
0.19
CMe
0.022
a
IC50 for inhibition of p38 kinase enzymatic activity. Data were generated either
by ELISA using commercially available human p38 and myelin basic protein as a
substrate, or using an Invitrogen Z’lyte kit with its proprietary substrate. IC50’s given
represent the means of a minimum of two, and generally three or more, indepen-
dent experiments. Standard deviation for assays typically 30% of the mean or less.
Table 4
Comparison of IC50 data of selected compounds for the biochemical- and phospho-
p38 assay
a
a
Compound
P38
a
IC50
(lM)
PP38-aIC50 (lM)
BIRB-796
SB203580
3
9
10
13
17
18
20
21
0.044
0.039
0.25
0.059
0.071
0.19
0.017
6.0
0.31
0.038
0.14
0.090
0.001
0.002
0.021
0.013
ATP Binding Site
Phe169 Pocket
Met (109)
H2N
Asp (168)
N
NR1R2
n
NH2
Polar/Hydrophobic
O
Pocket
0.072
0.025
MeO2SHN
N
H
0.008b
0.038
OMe
O
-
CO2
NH2
Glu71 Side Chain
Kinase Specificity
Pocket
a
Hydrophobic Pocket
Glu (71)
IC50’s given represent the means of a minimum of two, and generally three or
more, independent experiments. Standard deviation for assays typically 30% of the
Lys (53)
mean or less.
IC50 of LPS-stimulated TNF-
monocytic lineage (THP-1).
Figure 2. Simplified proposed binding mode of sulfonamide six-membered-ring
-ketoamides with directly linked rigid-spacers/pharmacophores in the ATP
binding site.
b
a production in immortalized human cells of a
a
R2
R2
Y
N
Y
N
HN
Br
Br
H2N
28a
ii
i
17-20
vi
O
O
26
O
O
R
OMe
27
29 R = OMe
30 R = OH
iv
Y
X
Y
X
R1
NH2
OMe
14
SnBu3
28b
22-25
vi
iii
O
O
R
31 R = OMe
32 R = OH
v
Scheme 2. Reagents and conditions: (i) CH2Cl2, AlCl3, rt then methyl oxalyl chloride, 16 h, 37%; (ii) Pd(OAc)2, BINAP, CsCO3, toluene, 100 °C, 24 h; (iii) Pd(PPh3)4, dioxane,
microwave, 150 °C, 10 min; (iv) CH2Cl2, BBr3, 0 °C, 35 min; (v) EtOAc, 40% aq KF, 1 h; (vi) PyBOP, HOBt, DIEA, DMF, rt, 16 h.

A. G. Montalban et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4819–4824
4823
X
Cl
N
R
N
N
N
MeO2SHN
NH2
O
Y
OH
OH
Cl
OMe
14
35
i
Br
21
vii
ii
O
O
33
O
O
OMe
36 X = Cl,
Y = Br, R = OMe
O, Y = Br, R = OMe
34
iii
iv
v
37 X =
38 X =
39 X =
40 X =
N
, Y = H, R = OMe
O
, Y = H, R = OH
O
, Y = H, R = Cl
O
N
N
N
vi
Scheme 3. Reagents and conditions: (i) CH2Cl2, AlCl3, rt then methyl oxalyl chloride, 2 h, 38%; (ii) acetone, K2CO3, 60 °C, 5 h, 75%; (iii) THF/toluene, morpholine, DIEA, 80 °C,
16 h, 80%; (iv) MeOH/CH2Cl2, 10% Pd/C, H2 (1 bar), rt, 50 min, 67%; (v) CH2Cl2, BBr3, 0 °C to rt, 10 min; (vi) CH2Cl2, DMF (cat), oxalyl chloride, rt, 2 h; (vii) CH2Cl2, DIEA, rt, 12 h.
Table 5
gionsof theprotein. Itappears, however, thatbymovingthenitrogen
Single dose plasma pharmacokinetic parameters of 6 and 9 following dosing in rats
further away from the pyridine ring, the protein can adjust slightly
so that the hydrogen bond acceptor moiety can form a good hydro-
gen bond interaction within the hinge region while maintaining
the favorable binding interaction of the sulfonamide with the argi-
Compound
6
9
Dose (mg/kg) iv
Dose (mg/kg) po
Tmax (h)
10
30
0.5
2.4
9.8
15
10
30
3.0
1.3
3.5
14
nine-rich area of p38a. The proposed allosteric binding mode is
Cmax
(l
g/mL)
shown in Figure 2 and generally consistent with data from our p38
T
½
(h)
phosphorylation inhibition assay.14 Table 4 summarizes some of
AUC (lg?h/mL)
the data which suggest that this new family of
a-ketoamides can
po F%
57
77
cause the necessary conformational change (DFG-in?DFG-out) to
inhibit phosphorylation of p38 by its upstream kinase (MKK3/
MKK6)15 similar to BIRB-7962 and our first and second generation
series.10,11 BIRB-796 and SB203580 are included for Ref. 10.
Ketoamides 17–20 and 22–25 were synthesized, as shown in
Scheme 2, through reaction of the respective anisidine derivatives
14 (R1 = NHSO2Me16 or CN) with the ketoacids 30 and 32. The keto-
acids 30 and 32, on the other hand, were prepared through palla-
dium catalyzed cross-coupling reaction of the correspondingly
substituted amino-pyrimidines/pyridines 28a and 28b17 with the
bromo derivative 27 after subsequent hydrolysis of the ketoester
intermediates 29 and 31, respectively. Compound 27 was in turn
obtained via acylation of commercially available bromonaphthol
26.
Compounds with an oxygen spacer linking the naphthyl- to the
six-membered-heteroaromatic-ring were generally prepared as
shown in the reaction sequence bellow for the final compound
21 (Scheme 3). Thus, acylation of naphthol 33 to give 34, followed
by nucleophilic substitution of the commercially available pyrimi-
dine derivative 35 gave the ketoester 36. Compound 36 was in turn
reacted with morpholine to give 37 which after reductive dehalo-
genation (38), hydrolysis (39) and treatment with oxalyl chloride
gave the requisite acid chloride 40. In the final step, the acid chlo-
ride 40 was reacted with 14 to give 21.
Cl (L/h/kg)
Vdss (L/kg)
1.1
4.9
1.6
5.2
properties consistent with drug like characteristics. This was fur-
ther confirmed by in vivo rat snapshot PK studies for which two
examples are given in Table 5. In addition, in vivo anti-inflamma-
tory activity for this class of p38 inhibitors was confirmed using
a mouse contact hypersensitivity model18 and in mouse and rat
models of acute and chronic inflammation.14
In summary, we have optimized a novel class of potent p38
inhibitors based on an
an orally available compound that is currently undergoing phase
II clinical trials in inflammatory conditions. We established SAR
trends that are consistent with the related BIRB-796 series and
achieved potencies in the double digit nanomolar range in both
a-ketoamide scaffold that finally yielded
p38
ex vivo inhibition in human whole blood and anti-inflammatory
efficacy in animal models. We further believe, based on the above
results, computational modeling and a phospho-p38 inhibition
a and cytokine inhibition. In addition, we showed good TNF-
a
a
assay, that these compounds inhibit p38, at least partially, via an
allosteric mode rather than an orthosteric/competitive mode of
binding.
References and notes
Compounds exemplary for their whole blood and rodent PK
properties are shown. Potent proinflammatory cytokine inhibition
for this series was demonstrated in vitro using a THP-1 whole cell
assay and ex vivo in rat and human whole blood. Compounds 19
1. Boehm, J. C.; Smietana, J. M.; Sorenson, M. E.; Garigipati, R. S.; Gallagher, T. F.;
Sheldrake, P. L.; Bradbeer, J.; Badger, A. M.; Laydon, J. T.; Lee, J. C.; Hillegass, L.
M.; Griswold, D. E.; Breton, J. J.; Chabot-Fletcher, M. C.; Adams, J. L. J. Med. Chem.
1996, 39, 3929.
2. (a) Regan, J.; Breitfelder, S.; Cirillo, P.; Gilmore, T.; Graham, A. G.; Hickey, E.;
Klaus, B.; Madwed, J.; Moriak, M.; Moss, N.; Pargellis, C.; Pav, S.; Proto, A.;
Swinamer, A.; Tong, L.; Torcellini, C. J. Med. Chem. 2002, 45, 2994. and
references therein; (b) Cirillo, P. F.; Hickey, E. R.; Moss, N.; Breitfelder, S.;
Betageri, R.; Fadra, T.; Gaenzler, F.; Gilmore, T.; Goldberg, D. R.; Kamhi, V.;
Kirrane, T.; Kroe, R. R.; Madwed, J.; Moriak, M.; Netherton, M.; Pargellis, C. A.;
and 20, for example, inhibited TNF-a production in vitro with an
IC50 = 8 nM, and ex vivo in human whole blood with an
IC50 = 150 and 100 nM, respectively.
In vitro ADME evaluation of this new series of
a-ketoamides
yielded plasma stability, HLM, CYP, Caco-2, and kinetic solubility

4824
A. G. Montalban et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4819–4824
Patel, U. R.; Qian, K. C.; Sharma, R.; Sun, S.; Swinamer, A.; Torcellini, C.;
Takahashi, H.; Tsang, M.; Xiong, Z. Bioorg. Med. Chem. Lett. 2009, 19, 2386. and
references therein.
and pdb1kv2.ent) and SB203580 (pdb1au9.ent) respectively, using the docking
Software GLIDE (v4.0, Schrödinger, Portland, OR).
13. (a) A refined receptor model of p38 MAP kinase, which includes the activation
loop missing in the X-ray crystal structure (pdb1kv2.ent) bound to BIRB796,
was built with the PRIME comparative modeling software package (v.1.2,
Schrödinger, LLC, New York, USA) using the above X-ray crystal structure as a
template. Compound 10 was then docked using the induced fit protocol (GLIDE
v4.0 and PRIME v1.5, Schrödinger, LLC, New York, USA).; (b) Sherman, W.; Day,
T.; Jacobson, M. P.; Friesner, R. A.; Farid, R. J. Med. Chem. 2006, 49, 534.
14. Garrido Montalban, A.; Boman, E.; Chang, C.-D.; Conde Ceide, S.; Dahl, R.;
Dalesandro, D.; Delaet, N. G. J.; Erb, E.; Ernst, J.; Gibbs, A.; Kahl, J.; Kessler, L.;
Lundström, J.; Miller, S.; Nakanishi, H.; Roberts, E.; Saiah, E.; Sullivan, R.; Wang,
Z.; Larson, C. J. Eur. J. Pharmacol. 2010, 632, 93.
3. Wrobleski, S. T.; Doweyko, A. M. Curr. Top. Med. Chem. 2005, 5, 1005. and
references therein.
4. (a) Jarvis, B.; Faulds, D. Drugs 1999, 57, 945; (b) Seymour, H. E.; Worsley, A.;
Smith, J. M.; Thomas, S. H. L. Br. J. Clin. Pharmacol. 2001, 51, 201; (c) Rutgeerts, P.
J. Aliment. Pharmacol. Ther. 1999, 13, 9; (d) Klinkhoff, A. Drugs 2004, 64, 1267;
(e) Barry, J.; Kirby, B. Expert Opin. Biol. Ther. 2004, 4, 975.
5. (a) Lee, J. C.; Kumar, S.; Griswold, D. E.; Underwood, D. C.; Votta, B. J.; Adams, J.
L. Immunopharmacology 2000, 47, 185; (b) Pargellis, C.; Regan, J. Curr. Opin.
Invest. Drugs 2003, 4, 566; (c) Kumar, S.; Boehm, J.; Lee, J. C. Nat. Rev. Drug Disc.
2003, 2, 717.
6. (a) Chakravarty, S.; Dugar, S. Annu. Rep. Med. Chem. 2002, 37, 177; (b) Adams, J.
L.; Badger, A. M.; Kumar, S.; Lee, J. C. Prog. Med. Chem. 2001, 38, 1.
7. Dominguez, C.; Powers, D. A.; Tamayo, N. Curr. Opin. Drug Disc Dev. 2005, 8, 421.
8. Fischer, P. M. Curr. Med. Chem. 2004, 11, 1563.
15. Sullivan, J. E.; Holdgate, G. A.; Campbell, D.; Timms, D.; Gerhardt, S.; Breed, J.;
Breeze, A. L.; Bermingham, A.; Pauptit, R. A.; Norman, R. A.; Embrey, K. J.; Read,
J.; VanScyoc, W. S.; Ward, W. H. J. Biochemistry 2005, 44, 16475. and references
therein.
9. Medicinal Chemistry Principles and Practice; King, F. D., Ed., 2nd ed.; The Royal
Society of Chemistry, 2002; p 76.
10. Garrido Montalban, A.; Boman, E.; Chang, C.-D.; Conde Ceide, S.; Dahl, R.;
Dalesandro, D.; Delaet, N. G. J.; Erb, E.; Ernst, J. T.; Gibbs, A.; Kahl, J.; Kessler, L.;
Lundström, J.; Miller, S.; Nakanishi, H.; Roberts, E.; Saiah, E.; Sullivan, R.; Wang,
Z.; Larson, C. J. Bioorg. Med. Chem. Lett. 2008, 18, 1772.
11. Garrido Montalban, A.; Boman, E.; Chang, C.-D.; Conde Ceide, S.; Dahl, R.;
Dalesandro, D.; Delaet, N. G. J.; Erb, E.; Gibbs, A.; Kahl, J.; Kessler, L.; Lundström,
J.; Miller, S.; Nakanishi, H.; Roberts, E.; Saiah, E.; Sullivan, R.; Wang, Z.; Larson,
C. J. Bioorg. Med. Chem. Lett. 2008, 18, 5456.
16. Goldberg, D. R.; Hao, M.-H.; Qian, K. C.; Swinamer, A. D.; Gao, D. A.; Xiong, Z.;
Sarko, C.; Berry, A.; Lord, J.; Magolda, R. L.; Fadra, T.; Kroe, R. R.; Kukulka, A.;
Madwed, J. B.; Martin, L.; Pargellis, C.; Skow, D.; Song, J. J.; Tan, Z.; Torcellini, C.
A.; Zimmitti, C. S.; Yee, N. K.; Moss, N. J. Med. Chem. 2007, 50, 4016.
17. Moss, N.; Breitfelder, S.; Betageri, R.; Cirillo, P. F.; Fadra, T.; Hickey, E. R.;
Kirrane, T.; Kroe, R. R.; Madwed, J.; Nelson, R. M.; Pargellis, C. A.; Qian, K. C.;
Regan, J.; Swinamer, A.; Torcellini, C. Bioorg. Med. Chem. Lett. 2007, 17, 4242.
18. Takanami-Onishi, Y.; Amano, S.; Kimura, S.; Asada, S.; Utani, A.; Maruyama, M.;
Osada, H.; Tsunoda, H.; Irukayama-Tomobe, Y.; Goto, K.; Karin, M.; Sudo, T.;
Kasuya, Y. J. Biol. Chem. 2002, 277, 37896.
12. Models of p38 in the DFG-out and DFG-in conformations were generated from
the X-ray crystal structures of BIRB-796 and one of its analogues (pdb1kv1.ent