Targeting the hinge glycine flip and the activation loop: Novel approach to potent p38α inhibitors

Article abstract of DOI:10.1021/jm300951u

The p38 MAP kinase is a key player in signaling pathways regulating the biosynthesis of inflammatory cytokines. Small molecule p38 inhibitors suppress the production of these cytokines. Therefore p38 is a promising drug target for novel anti-inflammatory drugs. In this study, we report novel dibenzepinones, dibenzoxepines, and benzosuberones as p38α MAP kinase inhibitors. Previously reported dibenzepinones and dibenzoxepines were chemically modified by introduction of functional groups or removal of a phenyl ring. This should result in targeting of the hydrophobic region I, the "deep pocket", and the hinge glycine flip of the kinase. Potent inhibitors with IC₅₀ values in the single digit nanomolar range (up to 3 nM) were identified. Instead of targeting the "deep pocket" in the DFG-out conformation, interactions with the DFG-motif in the in-conformation could be observed by protein X-ray crystallography.

Full text of DOI:10.1021/jm300951u

Article
pubs.acs.org/jmc
Targeting the Hinge Glycine Flip and the Activation Loop: Novel
Approach to Potent p38α Inhibitors
Kathrin E. Martz,† Angelika Dorn,† Benjamin Baur,† Verena Schattel,† Marcia I. Goettert,†
́
Svenja C. Mayer-Wrangowski,‡ Daniel Rauh,‡ and Stefan A. Laufer*,†
†Institute of Pharmacy, Department of Pharmaceutical and Medicinal Chemistry, Eberhard-Karls-University Tubingen, Auf der
̈
Morgenstelle 8, 72076 Tubingen, Germany
̈
‡Faculty of Chemistry−Chemical Biology, Technische Universitat Dortmund, Otto-Hahn-Straße 6, D-44227 Dortmund, Germany
̈
S
* Supporting Information
ABSTRACT: The p38 MAP kinase is a key player in signaling
pathways regulating the biosynthesis of inflammatory
cytokines. Small molecule p38 inhibitors suppress the
production of these cytokines. Therefore p38 is a promising
drug target for novel anti-inflammatory drugs. In this study, we
report novel dibenzepinones, dibenzoxepines, and benzosuber-
ones as p38α MAP kinase inhibitors. Previously reported
dibenzepinones and dibenzoxepines were chemically modified
by introduction of functional groups or removal of a phenyl
ring. This should result in targeting of the hydrophobic region
I, the “deep pocket”, and the hinge glycine flip of the kinase. Potent inhibitors with IC50 values in the single digit nanomolar
range (up to 3 nM) were identified. Instead of targeting the “deep pocket” in the DFG-out conformation, interactions with the
DFG-motif in the in-conformation could be observed by protein X-ray crystallography.
INTRODUCTION
■
Elevated levels of the proinflammatory cytokines tumor
necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) are
implicated in the onset of various autoimmune diseases such as
rheumatoid arthritis, Crohn's disease, and psoriasis. Recently
the role of p38 in CNS disorders, mainly Alzheimer's disease,
was discovered.1 Clinical and commercial successes of bio-
logical agents that neutralize TNF-α (etanercept, infliximab,
adalimumab, certolizumabpegol, and golimumab2−4) or IL-1β
(anakinra5) confirmed the central role of these cytokines in the
inflammation process. The mitogen-activated protein (MAP)
kinase p38 is a key player in signaling pathways regulating the
biosynthesis of inflammatory cytokines, including TNF-α and
IL-1β, at the translational and transcriptional levels.6 Small-
molecule inhibitors of p38 MAP kinase7−10 (Figure 1) such as
SB20358011,12 (1) can suppress the production of these
cytokines. This strategy seems to be promising because a
p38γ inhibitor was approved recently for idiopathic pulmonary
fibrosis.13 However, no p38α MAP kinase inhibitor made it to
the market yet, although there are two candidates in clinical
trials at the moment: PH-79780414−17 is in phase III for the
treatment of COPD, and BMS-58294918−20 is in phase II for
the treatment of rheumatoid arthritis. Therefore, there is a
continuous need for novel orally bioavailable small molecule
inhibitors of p38α MAP kinase.12,21,22
Figure 1. Inhibitors of p38 MAP kinase: pyridinylimidazole
SB20358011,12 (1) and biphenyl amide (2).15
the suberone scaffold. The linear constitution of the inhibitor is
tolerated due to the relatively small gatekeeper Thr106 in p38α
and contrasts with most other kinases that have more sterically
demanding residues at this position. The X-ray cocrystal
structure of the dibenzepinone inhibitor 3 in complex with
p38α24 shows that upon binding of the dibenzepinone inhibitor
3, the p38 MAP kinase undergoes a Gly110 flip. This glycine
flip is a small conformational rearrangement in the hinge region
of the p38α MAP kinase induced by the inhibitor. Gly110,
which is able to take a positive torsion angle, is switched about
180°, and 3 binds with its carbonyl oxygen toward the
backbone NH of Gly110 and the backbone NH of Met109.
We previously reported a novel series of dibenzepinone
inhibitors23 (Figure 2) belonging to the class of so-called linear
binders.8 The phenyl substituent of 3 occupies the hydrophobic
region I of the active site and is linked with an amino group to
Received: July 4, 2012
Published: August 16, 2012
© 2012 American Chemical Society
7862
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
Figure 2. Schematic drawing of important interactions between the competitive inhibitor 3 and the ATP binding site of p38α MAPK.
a
Scheme 1. Synthetic Route for the Dibenzepinone Scaffold 7
a
Reagents and conditions: (a) 1. triphenylphosphine, 3-chlorobenzylchloride, methanol, reflux; 2. 2-formylbenzoic acid, sodium methoxide, 0 °C;
(b) H2, Pd/C, ethyl acetate/acetonitrile, rt; (c) 1. oxalyl chloride, dimethylformamide, dichloromethane, rt; 2. AlCl3, dichloromethane, rt.
The glycine flip provides selectivity for p38α over other kinases
with less flexible, nonglycine residues at this position. Only 46
of all 518 kinases are able to perform a glycine flip. Of these 46
kinases, only 3 kinases possess a small threonine as gatekeeper
residue in addition to the glycine in the hinge region, which
could be utilized to gain further selectivity. In addition, the X-
ray structure provided the following information: the A-ring of
the dibenzepinone scaffold does not form any relevant
interactions with the enzyme and may therefore not be
essential for the inhibitory activity. In this study, we report on
the synthesis of novel p38α MAP kinase inhibitors based on
SAR from compound 3.
In a first step (Scheme 2), we focused on the variation of the
substitution pattern at the C-ring of the scaffold to target the
deep pocket9 of the p38α MAP kinase in order to achieve
additional interactions with the enzyme, improving the potency
of the lead compounds based on data from in vitro assays
(Tables 1−3).
Second, we removed the A-ring to get a benzosuberone
scaffold25 (Scheme 2) that showed similar inhibitory potency as
the dibenzepinone scaffold.26 Furthermore, we introduced an
oxygen to the suberone bridge to obtain the respective
dibenzoxepine derivative.27
7863
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthetic Route for the Benzosuberone Scaffold 11
a
Reagents and conditions: (a) 1. (3-carboxypropyl)triphenylphosphonium bromide, sodium methoxide, methanol, reflux; 2. 3-chlorobenzaldehyde,
methanol, reflux; (b) H2, Pd/C, ethyl acetate, rt; (c) 1. oxalyl chloride, dimethylformamide, dichloromethane, rt; 2. AlCl3, dichloromethane, rt.
Preparation of target compounds 17−40, 41−72, and 73−88
CHEMISTRY
■
The dibenzepinone scaffold26 was synthesized as shown in
Scheme 1: Wittig reaction of 2-carboxybenzaldehyde and 3-
chlorobenzylchloride in the presence of triphenylphosphine
and sodium methoxide yields the stilbeno-2-carboxylic acid 5.
The ethano linker 6 was obtained by reduction of the alkene 5
with H2 over Pd/C in ethyl acetate at room temperature. The
subsequent ring closure leading to the dibenzepinone 726 was
best achieved via an intramolecular Friedel−Crafts acylation
under standard conditions.
was conducted as shown in Schemes 4 and 5 by reaction of the
Scheme 4. Synthetic Route for Compounds 17−37, 41−68,
a
and 73−86
The benzosuberone scaffold was synthesized as shown in
Scheme 2.25 Wittig reaction of 3-chlorobenzaldehyde and (3-
carboxypropyl)triphenylphosphonium bromide in the presence
of sodium methoxide yields the unsaturated carboxylic acid 9.
The saturated carboxylic acid 10 was obtained by reduction of
the double bond of 9 with H2 over Pd/C in ethyl acetate at
room temperature. The subsequent ring closure to the
benzosuberone 1125 was best performed by an intramolecular
Friedel−Crafts reaction under standard conditions.
The dibenzoxepine scaffold27 was synthesized as shown in
Scheme 3: A Wohl−Ziegler bromination in carbon tetra-
Scheme 3. Synthetic Route for the Dibenzoxepinone Scaffold
16
a
a
For the nature of R1 and R2, see Tables 1 and 3.
Scheme 5. Synthetic Route for the Inverse Amide
a
Compounds 37−40, 69−72, and 87−88
a
Reagents and conditions: (a) N-bromosuccinimide, azobis-
(isobutyronitrile), carbon tetrachloride, reflux; (b) 1. 3-chlorophenol,
K2CO3, acetone, 50 °C; 2. methyl-2-(bromomethyl)-3-methoxyben-
zoate, acetone, 70 °C; (c) KOH, methanol, reflux; (d) 1. oxalyl
chloride, dimethylformamide, dichloromethane, rt; 2. AlCl3, rt.
chloride with N-bromo-succinimide and azobis-
(isobutyronitrile) yields the brominated compound 13. In the
next step, compound 14 was obtained by Williamson
etherification with 3-chloro-phenol. Then the ester was
disrupted to obtain the free carboxylic acid derivative 15.
Intramolecular ring closure was achieved via Friedel−Crafts
acylation to obtain the dibenzepinone scaffold 16.27
a
For the nature of R1 and R2, see Table 2.
7864
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
Figure 3. Chemical structure of compounds 18 and 45.
Figure 4. Docking result for compound 18. The hydrogen bonds are indicated as dotted lines.
Figure 5. Docking result for compound 45. The hydrogen bonds are indicated as dotted lines.
respective amine with either the dibenzepinone, dibenzoxepine,
or benzosuberone scaffold.29 The respective scaffold 7, 11, or
16 undergoes a palladium-catalyzed amination reaction with the
corresponding aryl amine resulting in the final compounds 17−
40, 41−72, and 73−88. Synthesis of the final compounds was
attained as described above using the respective amines as
coupling reagent for the Buchwald−Hartwig reaction.
test compounds to compete with ATP for the ATP binding site
of the kinase inversely correlates with the capacity of p38α
MAP kinase to phosphorylate its natural substrate, activating
transcription factor 2 (ATF-2), when incubated with ATP and
the test compound. SB20358011,12 was used as a reference
compound, and the test was performed with an optimized ATP
concentration of 100 μM.
BIOLOGICAL TESTING
BIOLOGICAL RESULTS AND DISCUSSION
■
■
The inhibitory potencies of compounds 17−40, 41−72, and
73−88 were evaluated using an isolated p38α enzyme assay,30
at concentrations ranging from 10 μM to 10 nM. The ability of
Docking Studies. Docking studies conducted with Glide
from Schroedinger28 suggested the meta-position of the C-ring
to be a good site for insertion of a benzamide moiety to achieve
7865
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
Figure 6. Crystal structure of p38α in complex with 45. The inhibitor binds to the ATP-pocket of p38α by forming hydrogen bonds (red dotted
lines) to the peptide backbone of the hinge region (light pink) (Met109, Gly110) and additional hydrogen bonds to Glu71 in the α-helix C (blue)
and Asp168 in the DFG-motif (orange). The DFG-motif adopts a DFG-in conformation. Electron density map (2Fo − Fc) is contoured at 1σ using
PyMOL (http://www.pymol.org).
additional interaction opportunities. The binding mode of
compound 18 (Figure 3) in p38α is shown in Figure 4.
Compound 45 (Figure 3) is supposed to interact with p38 in
the same way (Figure 5). The predicted binding mode could be
verified by X-ray crystallography of p38α in complex with
compound 45 (Figure 6). We intended to target the DFG-out
conformation. But the docking results already predicted
targeting of the DFG-motif located in the activation loop.
Variation of the Substitution Pattern of the C-Ring.
Because the C-ring addresses the hydrophobic region I to some
extent, we chose fluoro substituents at the 2- or 4-position
(compounds 19, 21, 43, 45, 75, and 77) since compounds with
such a substitution pattern showed favorable results in our
previously published series.23 For comparison, we further
attached a methyl group at the 2- or 4-position (compounds 18,
20, 42, 44, 74, and 76). For the biphenyl amide series,
exemplified by 2 (Figure 1), the methyl group on the toluene
ring is one of the key requirements for p38α inhibitory
activity.31 An analogue of 2 that lacks the 4-methyl group (also
called the “magic methyl” group)32,33 shows a nearly 100-fold
decreased p38α activity.31 Other p38α-inhibitors such as the 2-
tolyl-(1,2,3-triazol-1-yl-4-carboxamide) inhibitors from Boeh-
ringer Ingelheim34 and the pyrrolo[2,1-f][1,2,4]triazine inhib-
itors from Bristol-Myers Squibb35 also bear a methyl group. In
the dibenzepinone series (Table 1), the replacement of
hydrogen (compound 17) by a methyl group (compound
18) did not increase p38α activity, because compound 17
already showed an IC50 value of 36 nM. In fact, we found
another relevant position; compound 20, bearing a methyl
substituent at the 2-position, showed an IC50 value of 10 nM,
which resulted in a 4-fold higher inhibitory activity compared
with reference 3. Compound 21, carrying a fluoro substituent in
2-position, showed even higher activity against p38α MAP
kinase resulting in an IC50 value of 5 nM. However, the
evaluation of the benzosuberone series (Table 1) showed the
inverse effect: replacing the hydrogen at position 2 of
compound 41 with either fluorine (compound 45) or methyl
(compound 44), which proved to be important for the
dibenzepinone series, showed no effect or even decreased the
activity nearly 4-fold. For this series, we could confirm the
“magic” position described in the literature: substituting the
hydrogen at position 4 increased the inhibitory potency nearly
three times. This resulted in IC50 values as low as 12 nM for
compound 43, bearing a fluoro substituent at position 4.
However, the “magic methyl” effect is not as pronounced as
described in the literature, which could be explained by the
different binding mode of the compounds. As described below,
they do not interact with the deep pocket but with the
activation loop. In the dibenzoxepine series, position 2 as well
as position 4 substituted derivatives (compounds 74−77)
showed an increased inhibitory activity, compared with
compound 73, up to an IC50 value of 17 nM for compound 74.
Inverse Amide. In order to verify whether the regiochem-
istry of the amide is important for the inhibitory activity, we
synthesized compounds 37−40, 69−72, and 87−88 (Table 2).
Kinase assay data showed that the inverse amide moiety results
in much poorer inhibitory activity (compare 17, 41, and 73
with 37, 69, and 87). Thus, the variety of the amide plays a
crucial role in the interaction between inhibitor and p38α.
Investigators from GlaxoSmithKline obtained opposite results:
compared with compound 2, the compound modified with
another type of amide (Figure 7) showed poorer inhibitory
activity, with IC50 values of 2.3 μM, compared with 0.075 μM
for 2. Additional studies with other p38α inhibitors such as 2-
tolyl-(1,2,3-triazol-1-yl-4-carboxamide) inhibitors from Boeh-
ringer Ingelheim34 and pyrrolo[2,1-f][1,2,4]triazine inhibitors
7866
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
Table 1. Evaluation of p38α MAP Kinase Activity for Compounds 17−21, 41−45, and 73−77 Based on the Rate of
c
Phosphorylation of ATF-2 (Activation Transcription Factor 2) Using an Isolated p38α Assay13
a
b
Mean
SD of three trials. Substituents are numbered as shown for presentational uniformity; see Experimental Section and Supporting
c
Information for correct numbering of substituents. IC50 of SB203580 = 0.037 0.003 μM (n = 25).
from Bristol-Myers Squibb35 have reported compounds with
only one type of amide. The structures of inhibitors from
Bristol-Myers Squibb35 are comparable to our most favorable
substituent. In comparison to our most successful compounds,
Boehringer Ingelheim34 reported inhibitors with the inverse
structure of the amide. This could be explained by a slightly
different position of the inhibitor in complex with the enzyme
due to either one or two hydrogen bonds between inhibitor and
hinge region.
Variation of the D-Ring. Compounds 22 and 46−48,
bearing a tert-butyl moiety in the para-position of the D-ring,
were not tolerated as well as the extended acetic and propionic
acid moieties (29−30 and 58−61), presumably due to steric
hindrance (Table 3). This could also explain the decreased
activity of compounds 50 and 52 in comparison to compound
24.
Based on docking studies, we chose to vary the meta-position
in the substitution pattern of the D-ring of the dibenzepinone,
the dibenzoxepine, and the benzosuberone scaffold. Com-
pounds with various substituents 23, 49, and 78, 24 and 51,
and 25−28, 53−57 and 79 and 80 were tested in the enzyme
assay. Besides compound 26, compounds 24 and 28 resulted in
comparable or even better inhibitory activity than the
unsubstituted compound 17. The introduction of a methoxy
group (compound 24) in the meta-position improved the
inhibitory activity by about 3-fold. This outcome could be due
to a direct interaction between the methoxy group and the
enzyme. Interestingly, the fluoro substituent in the para-
position (26) yielded a 10-fold higher inhibitory activity
compared with the meta-position (25). Compound 26 was
similar in inhibitory activity in comparison to compound 17
with the unsubstituted D-ring. Unexpectedly, this structure−
activity relationship only applies to the dibenzepinone scaffold,
not to the benzosuberone scaffold. The comparison of
compound 51 bearing the meta-methoxy group with the
unsubstituted compound 41 shows a 4-fold decreased
inhibitory activity. Compound 52 with a para-methoxy
substituent showed the best activity in the methoxy series
with an IC50 of 84 nM. Combination of the 2-fluoro substituent
on the C-ring with the meta-fluoro substituent on the D-ring
resulted in a good inhibitory activity (compound 54, IC50 = 40
nM). In the dibenzoxepine series, two substituents in the meta-
position with good inhibitory activity could be identified:
compound 78 with a meta-methyl substituent (IC50 = 15 nM)
and compound 80 with a meta-morpholino substituent (IC50
=
30 nM). These results indicate that the three scaffolds are not
comparable at all regarding their inhibitory effects.
Results of the replacement of the phenyl by various
heterocycles are summarized in Table 3. The aromatic
heterocycles (32−36, 63−67, and 81−85) seem to be
7867
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
Table 2. Evaluation of p38α MAP Kinase Activity Based on the Rate of Phosphorylation of ATF-2 Using an Isolated p38α
b
Assay13
a
b
Mean SD of three trials. IC50 of SB203580 = 0.037 0.003 μM (n = 25).
compound of this series is compound 68, which combines
two substituents on the C-ring with the 3-thiophene as
replacement for the phenyl ring D. This results in an excellent
inhibitory activity of 3 nM.
Kinase Selectivity Profiling. As we removed the A-ring of
the dibenzepinone scaffold and introduced amide residues to
the C-ring, it was necessary to confirm that this did not lead to
a loss of selectivity in comparison to the highly selective
reference compound. Therefore selectivity profiling of com-
pound 45 against a panel of 333 kinases was performed.36 At a
concentration of 10 μM, only 6 out of 333 kinases were
inhibited >50%: CAMK2D, CK1-δ, CK1-ε, SLK, p38α, and
p38β. However, only p38α and p38β were inhibited
substantially. This leads to the result that the loss of the A-
ring and the introduction of amide moieties do not affect the
selectivity.
Physicochemical Properties. The physicochemical prop-
erties of all the compounds are not perfect: the log P is rather
high with values over 5. This may result in some drawbacks
such as low solubility, which is a common problem of this
amide series. Thus there is further work necessary to improve
the ADME properties of the amide compounds.
X-ray Crystallography. The binding mode of compound
45 in complex with p38α MAP kinase (Figure 6) was
determined by protein X-ray crystallography (Table 4). The
central carbonyl oxygen of the benzosuberone scaffold induces
the expected glycine flip in the hinge region and forms two
hydrogen bonds with the backbone of Met109 and Gly110. In
addition, polar interactions can be observed between the amide
NH and Glu71 (α-helix C) and the amide carbonyl oxygen and
Figure 7. Inverse amide compound from GlaxoSmithKline.31
advantageous compared with the cyclopropyl moiety (31 and
62). The location of the heteroatom in meta-position (32−36,
63−67, and 81−85) was more favorable than in ortho-position.
This effect confirms the result of compound 24: the
introduction of a hydrogen bond acceptor in the meta-position
ameliorates the inhibitory effect, probably due to an additional
interaction opportunity. Interestingly, the location of the
heteroatom plays a crucial role for the benzosuberone scaffold
as well. At first sight, there seems to be no explanation for the
poor effect of the meta-substituents compared with the
promising results from the heterocycles.
However, all the differences in the structure−activity
relationships described above cannot be explained by the
docking results.
Finally, the combination of the best substituents of the C-
ring and the best variation of the D-ring resulted in compound
83 with a low nanomolar inhibitory activity (IC50 = 10 nM) and
in compound 34, where the combination of the best
substitution pattern resulted in an increase of potency; the
compound showed an IC50 of 4 nM. The most active
7868
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
Table 3. Evaluation of p38α MAP Kinase Activity for Compounds 22−36, 46−68, and 78−86 Based on the Rate of
c
Phosphorylation of ATF-2 Using an Isolated p38α Assay13
a
b
c
Mean SD of three trials. % Inhibition at 10 μM. See Experimental Section/Supporting Information for correct numbering of substituents. IC50
of SB203580 = 0.037 0.003 μM (n = 25).
Asp168 in the DFG-motif of the activation loop. The
phenylamino moiety of the inhibitor occupies the hydrophobic
region I of the kinase to some extent, and the kinase adopts the
active DFG-in conformation as observed for common type I
inhibitors like Scio-496.37,38 Interestingly, some residual
electron density around the flexible activation loop indicates
partial existence of the DFG-out conformation although to a
much weaker extent. Partial or mixed occupancies in the
activation loop of protein kinases are not uncommon and
highlight that 45 can bind to both DFG conformations.
However, the phenylamino moiety of 45 is not bulky enough to
induce a complete shift to the DFG-out conformation as
observed for classic type II inhibitors such as BIRB-79639 and
others.40,41
7869
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
1
corrected. H NMR (200/400 MHz) and 13C NMR (50/100 MHz)
spectra were recorded on a Bruker Avance 200 and a Bruker Avance
400. Chemical shifts are reported in ppm from the solvent resonance.
IR data were determined on a Perkin-Elmer Spectrum One
spectrometer (ATR technique). Flash chromatography was performed
using a LaFlash system (VWR) with Merck silica gel (25−40 μm).
The purity of the final compounds was determined by HPLC (Merck
Hitachi L-6200 intelligent pump, Merck Hitachi AS-2000 autosampler,
Merck Hitachi L-4250 UV−vis detector) using a LiChrospher C18
column (5 μm), employing a gradient of 0.01 M KH2PO4 (pH 2.3)
and methanol as solvent system with a flow rate of 1.0 mL/min and
detection at 254 nm. Mass spectra were run on a Hewlett-Packard HP
6890 series GC-system equipped with a HP-5MS capillary column
(0.25 μM film thickness 30 m × 0.25 mm) and a Hewlett-Packard HP
5973 mass selective detector (70 eV). HRMS (EI) (electron impact−
high-resolution mass spectroscopy) data were obtained from the
department for mass spectrometry, Institute of Organic Chemistry,
Table 4. Data Collection and Refinement Statistics for p38α
with Compound 45 (3UVP)
Data Collection
space group
cell dimensions
a, b, c (Å)
P212121
64.00, 68.80, 74.80
90.00, 90.00, 90.00
50.0−2.4 (2.50−2.40)
9.3 (41.2)
α, β, γ (deg)
resolution (Å)
Rsym or Rmerge (%)
I/σI
a
15.43 (3.95)
completeness (%)
redundancy
99.8 (99.6)
5.19 (4.59)
Refinement
resolution (Å)
48.63−2.40
13405
no. reflns
Eberhard-Karls-University Tubingen. All compounds were >95% pure.
See Supporting Information for details.
̈
R
work/Rfree
no. atoms
protein
21.5/29.7
Amination Reaction with the Dibenzepinone Scaffold
(General Procedure A). A mixture of 2.1 mmol aryl chloride, 2.1
mmol amine, 4.6 mmol of the corresponding base, 90 mg (0.19 mmol)
of 2-(dicyclohexylphosphino)-2′,4′,6′-triisopropylbiphenyl and 20 mg
(0.09 mmol) of Pd(OAc)2 in 2 mL of absolute tert-butanol and 8 mL
of absolute toluol was stirred at 90 °C under an atmosphere of argon.
For compounds 17, 25, and 37, Cs2CO3 was used as base, for
compounds 20, 23, 27, 31, 33, 34, 35, 39, and 40, KOt-Bu was used as
base, and for the other compounds, NaOt-Bu was used. When the
reaction was completed, the mixture was allowed to cool to room
temperature, diluted with water, and extracted with ethyl acetate. The
extracts were combined, washed with saturated saline solution, and
then dried over Na2SO4. The solvent was removed under vacuum, and
the residue was purified by flash chromatography (SiO2 60, n-hexane/
ethyl acetate).
2692
69
ligand/ion
water
50
B-factors
37.2
37.3
38.3
31.7
protein
ligand/ion
water
rms deviations
bond lengths (Å)
bond angles (deg)
structure (PDB-ID code)
wavelength (Å)
0.014
1.488
p38α with 45 (3UVP)
0.978 600
temperature
90 K
X-ray source
SLS X10SA
Amination Reaction with the Benzosuberone Scaffold
(General Procedure B). A mixture of 1.85 mmol of aryl chloride,
1.85 mmol of amine, 22 mmol of the corresponding base, 110 mg (0.2
mmol) of 2-(dicyclohexylphosphino)-2′,4′,6′triisopropylbiphenyl, and
20 mg (0.09 mmol) of Pd(OAc)2 in 2 mL of absolute tert-butanol and
10 mL of absolute toluol was stirred at 90 °C under an atmosphere of
argon. For compounds 42, 45, 46, 47, 48, 53, 54, and 71, NaOt-Bu
was used as base, and for the other compounds, KOt-Bu was used.
When the reaction was completed, the mixture was allowed to cool to
room temperature. It was diluted with water and subsequently
extracted with diethyl ether. The extracts were combined, washed with
saturated saline solution, and afterwards dried over Na2SO4. The
solvent was removed under vacuum, and the residue was purified by
flash chromatography (SiO2 60, n-hexane/ethyl acetate).
Amination Reaction with the Dibenzoxepinone Scaffold
(General Procedure C). A mixture of 200 mg (0.81 mmol) of 3-
chloro-dibenzo[b,e]oxepin-11(6H)-one, 700 mg of Cs2CO3 (2.14
mmol), 250 mg (1.18 mmol) of amide, 100 mg (0.21 mmol) of 2-
(dicyclohexylphosphino)-2′,4′,6′-triisopropylbiphenyl, 20 mg (0.09
mmol) of Pd(OAc)2, 10 mL of absolute 1,4-dioxane and 2 mL of
absolute tert-butanol was heated under argon to 110 °C and stirred for
3 h. When the reaction was completed, the mixture was allowed to
cool to room temperature. Then the solvents were evaporated, and the
residue was purified by flash chromatography (SiO2 60, dichloro-
methane/ethanol).
2-(3-Chlorophenethenyl)benzoic acid (5). Triphenylphosphine
(15.3 g, 58 mmol) was diluted in 100 mL of methanol, 9.4 g (58
mmol) of 3-chlorobenzylchloride (diluted in 50 mL of methanol) was
added dropwise to the stirred solution, and the solution was heated
under reflux for 2 h. The reaction mixture was cooled to 0 °C, and 8.7
g (58 mmol) of 2-formylbenzoic acid was added. At 0 °C, 28.0 g (145
mmol) of sodium methoxide (28% in methanol) was added dropwise
over a period of 45 min. Afterwards the solution was stirred for 3 h at 0
°C. The reaction mixture was poured onto a stirred mixture of 75 g of
ice and 175 mL of H2O, filtered, and washed several times with H2O,
and the combined aqueous phases were subsequently washed several
Ramachandran plot
residues in most favored regions
87.7%
11.3%
1.0%
residues in additional allowed regions
residues in generously allowed regions
residues in disallowed regions
0.0%
a
Diffraction data from one crystal was used to determine the complex
structure. Values in parentheses refer to the highest resolution shell.
CONCLUSION
■
The removal of the A-ring had no substantial effect on the
inhibitory activity of p38 MAP kinase because the respective
benzosuberone derivatives showed comparable inhibitory
effects to the dibenzepinone and dibenzoxepine scaffolds.
By variation of the phenylamino moiety of the three scaffolds
and introduction of a substituted benzamide moiety that
afforded further interaction opportunities, we obtained
compounds with improved inhibitory activity. As suggested
by docking studies, the meta-position of the ring D proved to
be suitable for the introduction of further substituents. The
binding mode of compound 45 could be confirmed by X-ray
crystallography: compound 45 induces the hinge glycine flip
and binds to the activation loop. Compared with the lead
structure 3, the inhibitory potency could be increased by about
10-fold down to the low nanomolar range as exemplified by
compounds 21, 34, 68, and 84 with IC50 values of 5, 4, 3, and
10 nM, respectively.
EXPERIMENTAL SECTION
All commercially available reagents and solvents were used without
further purification. Melting points were determined with a Buchi
■
̈
melting point device, model B-545, and were thermodynamically
7870
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
times with dichloromethane. The product was precipitated by
reflux. Then 100 mg (0.60 mmol) of azobis(isobutyronitrile) was
added. After the reaction was completed, the solid was filtered, and the
organic phase was evaporated in vacuo to obtain a yellow oil. 1H NMR
(DMSO-d6) δ in ppm: 3.87 (s, 3 H, -OCH3), 5.01 (s, 2 H, -CH2Br),
7.46 (m, 1 H, C3H), 7.59 (m, 2 H, C5H, C4H), 7.87 (m, 1 H, C6H).
Methyl-2-[(3-chlorophenoxy)methyl]benzoate (14). K2CO3
(2.42 g, 17.40 mmol) was suspended in acetone. Then 2.24 g
(17.40 mmol) of 3-chlorophenol was added. After heating to 50 °C,
4.00 g (17.40 mmol) of methyl-2-(bromomethyl)-3-methoxybenzoate,
dissolved in acetone, was added, and the mixture was heated for 6 h at
70 °C. After the reaction was completed, the solvent was removed, and
the residue was dissolved in ethyl acetate. The organic layer was
washed with 10% HCl and then dried over Na2SO4. The extract was
further purified by flash chromatography (petrol ether/ethyl acetate,
3:1) to obtain a white oil. IR (ATR): 3403, 2961, 1711, 1432, 1265,
acidification of the aqueous extract. Mp: 125 °C. IR (ATR): 2816,
1
2648, 2530, 1673, 1567, 1416, 1305, 1273, 799, 784, 715 cm−1. H
NMR (DMSO-d6) δ in ppm: 6.59 (d, 1H, J = 12.3 Hz), 6.91−6.94 (m,
1H), 7.03−7.19 (m, 5 H), 7.35−7.42 (m, 2H).
2-(3-Chlorophenethyl)benzoic acid (6). Five grams (19 mmol)
of 5 was dissolved in a mixture of 75 mL of ethyl acetate and 75 mL of
acetonitrile, and 100 mg of palladium on activated carbon was added.
The mixture was stirred under hydrogen atmosphere for 5 h at room
temperature. The mixture was filtered, and the solvent was removed in
vacuo. Mp: 220 °C. IR (ATR): 2930, 1697, 1578, 1490, 1442, 1381,
1
1261, 1139, 1080, 988, 741 cm−1. H NMR (DMSO-d6) δ in ppm:
2.78−2.86 (m, 2 H), 3.13−3.21 (m, 2 H), 7.15−7.33 (m, 5 H), 7.43
(d, 1 H, J = 7.2 Hz), 7.83 (d, 1 H, J = 8.0 Hz), 12.95 (s, 1 H).
2-Chloro-10,11-dihydrodibenzo[a,d]cyclohepten-5-one (7).
Five grams (20 mmol) of 6 was dissolved in 150 mL of
dichloromethane, and then 2.54 g (20 mmol) of oxalyl chloride and
catalytic amounts of dimethylformamide were added dropwise, and the
mixture was stirred at room temperature for 1 h under argon
atmosphere. The mixture was added to a suspension of 3.4 g (30
mmol) of AlCl3 in dichloromethane and stirred for 4 h at room
temperature and afterwards poured onto ice, extracted with dichloro-
methane, washed with NaOH and H2O, and dried over Na2SO4, and
the solvent was removed in vacuo. Mp: 69 °C. IR (ATR): 2942−2847,
1767, 1634, 1581, 1283, 1242, 1186, 1105, 940, 911, 829, 760, 685
cm−1. 1H NMR (DMSO-d6) δ in ppm: 3.10 (s, 4 H), 7.27−7.49 (m, 5
H), 7.83−7.87 (m, 2 H).
1
1034, 857, 761, 739, 677 cm−1. H NMR (DMSO-d6) δ in ppm: 3.80
(s, 3 H, -OCH3), 5.42 (s, 2 H, CH2O), 6.99 (m, 3 H, C2′H, C4′H,
C6′H), 7.31 (m, 1 H, C3′H), 7.47 (m, 1 H, C2H,), 7.64 (m, 2 H, C4H,
C3H), 7.90 (m, 1 H, C6H).
[(3-Chlorophenoxy)methyl]benzoic acid (15). Methyl-2-[(3-
chlorophenoxy)methyl]benzoate (3.00 g, 10.87 mmol) was dissolved
in 40 mL of methanol and heated to 50 °C until the solution was clear.
Then 1.00 g (17.86 mmol) of KOH dissolved in 10 mL of H2O was
added, and the reaction mixture was refluxed for 4 h. The solvent was
then removed in vacuo, and the residue was dissolved in H2O. The
solution was then acidified under ice cooling with 10% HCl,
whereupon the product crystallized as a white solid. IR (ATR):
1
3343, 2970, 1587, 1565, 1272, 1231, 1113, 1039, 769, 738 cm−1. H
5-(3-Chlorophenyl)pent-4-enoic acid (9). To a suspension of
13.91 g (32.4 mmol) of (3-carboxypropyl)triphenylphosphonium
bromide in anhydrous methanol, 10.80 g (60 mmol) of 30% sodium
methoxide in methanol was added. The mixture was heated to reflux
under argon atmosphere for 1 h, and after that a solution of 3.94 g (28
mmol) of 3-chlorobenzaldehyde in methanol was dropwise added. The
resulting mixture was stirred for further 6 h, allowed to cool to room
temperature, poured onto ice, and extracted with diethyl ether.
Subsequently, the aqueous solution was acidified with concentrated
HCl and extracted three times with diethyl ether. The organic layer
was washed with H2O, dried over Na2SO4, and concentrated in vacuo.
The residue was purified using flash chromatography (petrol ether/
ethyl acetate, 1:1) to afford a yellowish oil. IR (ATR): 2941, 2866,
1727, 1675, 1589, 1282, 1259, 1091, 964, 824, 785, 751 cm−1. HPLC:
NMR (DMSO-d6) δ in ppm: 5.46 (s, 2 H, CH2O), 6.96 (d, J = 8.34
Hz, 1 H, C6′H), 7.01 (d, J = 8.08 Hz, 1 H, C4′H), 7.06 (s, 1 H, C2′H),
7.33 (m, 1 H, C5′H), 7.46 (m, 1 H, C3H), 7.61 (m, 2 H, C5H, C4H),
7.93 (d, J = 7.83 Hz, 1 H, C6H). 13C NMR (DMSO-d6) δ in ppm: 68.0
(CH2O), 113.7 (C6′), 114.8 (C2′), 120.7 (C4′), 127.8 (C5′), 128.1 C5),
129.7 (C1), 130.4 (C3), 130.9 (C4), 132.0 (C6), 133.7 (C2), 137.6
(C3′), 159.3 (C1′), 168.1 (CO2H).
3-Chlorodibenzo[b,e]oxepin-11(6H)-one (16). 2-(3-
Chlorophenoxy)methylbenzoic acid (2.0 g, 7.36 mmol) was dissolved
in 40 mL of dichloromethane and 1 mL of dimethylformamide and
treated with 0.97 g (7.36 mmol) of oxalyl chloride. After 30 min of
stirring, 2.30 g (17.30 mmol) of AlCl3 was added to the mixture. After
30 min, the reaction mixture was hydrolyzed with H2O, and the
organic layer was concentrated in vacuo. The extract was further
purified by flash chromatography (petrol ether/ethyl acetate, 3:1) to
obtain a white solid. IR (ATR): 3070, 1593, 1273, 1234, 1203, 1019,
874, 830, 756, 679 cm−1. 1H NMR (DMSO-d6) δ in ppm: 5.34 (s, 2 H,
C6H2), 7.22 (m, 2 H, C4H, C2H),7.56 (m, 2 H, C9H, C7H), 7.67 (m, 1
H, C8H), 7.79 (d, J = 7.58 Hz, 1 H, C1H), 8.10 (d, J = 8.59 Hz, 1 H,
C10H). 13C NMR (DMSO-d6) δ in ppm: 73.0 (C6), 120.2 (C4), 122.5
(C2), 123.9 (C11a), 128.4, (C7) 128.9 (C9), 129.3 (C10), 133.1 (C8),
133.2 (C1), 135.6 (C10a), 139.6 (C6a), 139.7 (C3), 161.2 (C4a), 189.1
(C11).
N-{[3-(5-Oxo-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-2-
yl)amino]phenyl}benzamide (17). Compound 17 was prepared
according to general procedure A. The residue was purified by flash
chromatography (n-hexane/ethyl acetate 1:1) to afford 145 mg (17%)
as a yellow solid. Mp: 189 °C. IR (ATR): 3308, 1643, 1596, 1545,
1480, 1333, 1269, 1108, 861, 759, 691 cm−1. HPLC: 9.5 min (99.7%).
1H NMR (DMSO-d6) δ in ppm: 3.08 (s, 4H), 6.93−7.06 (m, 3H),
7.28−7.59 (m, 8H), 7.80−8.03 (m, 5H), 8.90 (s, 1H, NH), 10.6 (s,
1H, NH). 13C NMR (DMSO-d6) δ in ppm: 34.4, 36.1, 111.5, 113.4,
114.3, 114.9, 115.3, 126.8, 127.9, 128.0 (×2), 128.7 (×2), 129.1, 129.7,
130.4, 131.9, 132.2, 133.7, 135.4, 139.4, 140.4, 141.8, 142.0, 145.7,
148.5, 166.0, 191.2. HRMS-ESI, m/z (C28H22N2O2): calcd, 418.1681
[M + H]+; found, 418.1663.
1
7.34 min (98.8%). H NMR (DMSO-d6) δ in ppm: 2.38 (s, 4 H),
6.38−6.39 (m, 2 H), 7.24−7.29 (m, 4 H).
5-(3-Chlorophenyl)pentanoic acid (10). Four grams (19 mmol)
of 9 was dissolved in 100 mL of ethyl acetate, and 400 mg of palladium
on activated carbon was added. The resulting suspension was stirred
under hydrogen atmosphere for 5 h at room temperature. Afterwards,
the mixture was filtered and concentrated in vacuo to afford an oil. 1H
NMR (DMSO-d6) δ in ppm: 1.45−1.57 (m, 4 H), 2.18−2.28 (m, 2
H), 3.85 (s, 2 H), 7.11−7.24 (m, 4 H).
2-Chloro-6,7,8,9-tetrahydrobenzocyclohepten-5-one (11).
Three grams (14 mmol) of 10 was dissolved in 20 mL of
dichloromethane, 1.77 g (14 mmol) of oxalyl chloride and catalytic
amounts of dimethylformamide were added dropwise, and the mixture
was stirred at room temperature for 1 h under argon atmosphere. The
mixture was added to a suspension of 9.0 g (367 mmol) of AlCl3 in
dichloromethane and stirred for 4 h at room temperature and then
poured onto ice, extracted with dichloromethane, washed with NaOH
and H2O, and dried over Na2SO4, and the solvent removed in vacuo.
The extract was further purified by flash chromatography (n-hexane/
ethyl acetate, 9:1) to obtain a yellowish oil. IR (ATR): 2920, 2612,
1703, 1592, 1575, 1472, 1419, 1288, 1255, 1203, 1076, 962, 887, 713
1
cm−1. HPLC: 7.17 min (99.1%). H NMR (DMSO-d6) δ in ppm:
1.64−1.80 (m, 4 H), 2.67 (t, 2 H, J1 = 5.56 Hz, J2 = 6.3 Hz), 2.92 (t, 2
H, J1 = 6.7 Hz, J2 = 5.56 Hz), 7.40−7.42 (m, 2 H), 7.56 (d, 1H, J = 8.2
Hz). 13C NMR (DMSO-d6) δ in ppm: 20.5, 24.8, 31.4, 40.7, 126.9,
129.8, 130.4, 137.1, 137.4, 144.1, 204.2.
Methyl-2-(bromomethyl)benzoate (13). Four grams (26.40
mmol) of methyl-2-methylbenzoate and 4.74 g (26.40 mmol) of N-
bromosuccinimide were dissolved in tetrachloromethane and heated to
N-[3-(5-Oxo-6,7,8,9-tetrahydro-5H-benzocyclohepten-2-
ylamino)phenyl]benzamide (41). Compound 41 was prepared
according to general procedure B. The residue was purified by flash
chromatography (n-hexane/ethyl acetate 3:2) to afford 118 mg
(17.3%) as a yellow solid. Mp: 190 °C. IR (ATR): 3316, 2939, 2860,
1639, 1610, 1575, 1527, 1481, 1276, 861, 824, 690 cm−1. HPLC: 8.11
7871
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
min (96.8%). 1H NMR (DMSO-d6) δ in ppm: 1.77 (m, 4 H), 2.62 (t,
2 H, J1 = 4.04 Hz, J2 = 6.44 Hz), 8.56 (t, 2 H, J1 = 5.92 Hz, J2 = 8.06
Hz), 6.91−7.01 (m, 3 H), 7.22−7.39 (m, 2 H), 7.48−7.62 (m, 4 H),
7.78 (s, 1 H), 7.92−7.96 (m, 2 H), 8.73 (s, 1 H, NH), 10.22 (s, 1 H,
NH). 13C NMR (DMSO-d6) δ in ppm: 20.6, 25.0, 32.6, 41.1, 111.2,
113.0, 114.0, 114.9, 115.7, 128.0, 128.7, 129.1 (2×), 129.7 (2×), 130.9,
131.9, 135.4, 140.4, 142.1, 144.6, 148.1, 166.0, 202.2. HRMS-ESI, m/z
(C24H22N2O2): calcd, 370.1681 [M + H]+; found, 370.1705.
AUTHOR INFORMATION
Corresponding Author
*E-mail: stefan.laufer@uni-tuebingen.de. Telephone: +49-
7071-2972459. Telefax: +49-7071-295037.
■
Notes
The authors declare no competing financial interest.
N-{3-[(11-Oxo-6,11-dihydrodibenzo[b,e]oxepin-3-yl)amino]-
phenyl}benzamide (73). Compound 73 was prepared according to
general procedure C. The residue was purified by flash chromatog-
raphy (dichloromethane/ethanol 95:5) to afford 45 mg (13.2%) as a
yellow solid. Mp: 97 °C. IR (ATR): 1588, 1562 1476, 1300, 1276,
1120, 871, 693 cm−1. HPLC: 7.92 min (100%). 1H NMR (DMSO-d6)
δ in ppm: 5.21 (s, 2 H, C6″H2), 6.65 (d, J = 2.02 Hz, 1 H, C4″H), 6.82
(d, J = 2.02 Hz, 1 H, C2′H), 6.92 (m, 1 H, C2″H), 7.30 (m, 1 H, C4′H),
7.51 (m, 7 H), 7.79 (m, 2 H, C5′H, C6′H, C7″H, C9″H, C1″H, C8″H,
C4H), 7.83 (m, 2 H, C3H, C5H), 7.94 (m, 2 H, C2H, C6H), 8.03 (d, J
= 8.97 Hz, 1 H, C10″H), 9.05 (s, NH, 1 H) 10.27 (s, NH, 1 H). 13C
NMR (DMSO-d6) δ in ppm: 73.1 (C6″), 102.7 (C4″), 110.9 (C2″),
111.9 (C6′), 114.8 (C2′), 115.8 (C4′), 117.1 (C11a), 128.0 (C2, C5),
128.4 (C7″), 128.7 (C3, C6), 129.2 (C9″), 129.3 (C5′), 129.8 (C10″),
131.9 (C8″), 132.6 (C1″), 133.7 (C4), 135.3 (C10a), 136.3 (C1), 140.5
(C1′), 140.5 (C3′), 141.3 (C6a″), 151.2 (C3″), 163.2 (C4a″), 166.0 (C
OAmide), 187.5 (C11″). HRMS-ESI, m/z (C27H20N2O3): calcd,
443.1366 [M + Na]+; found, 443.1364.
Crystallization and Structure Determination of p38α MAP
Kinase in Complex with Compound 45. Inactive (nonphosphory-
lated) human p38α MAP kinase was expressed and purified as
described previously.42 The purified protein was concentrated to 14
mg/mL and flash frozen in liquid nitrogen. Protein crystals of p38α
MAP kinase were obtained using crystallization conditions similar to
those previously reported.43 Briefly, apo-crystals of p38α were grown
in 24-well crystallization plates using the hanging-drop vapor diffusion
method at 20 °C by mixing 1 μL of protein solution (14 mg/mL) with
1 μL of reservoir solution (100 mM MES, pH 6.0−6.3, 22−27%
PEG4000, 50 mM n-octyl-β-D-glucopyranoside). For soaking experi-
ments, 20 μL of reservoir solution was mixed with 0.2 μL of
compound 45 (100 mM in DMSO) and centrifuged for 5 min at
13 000 rpm to remove excess compound. Apo-crystals were transferred
to a new drop (1 μL) of this solution and incubated at 20 °C for 24 h.
Crystals were cryoprotected by addition of 25% PEG400 and
subsequently flash frozen in liquid nitrogen.
ACKNOWLEDGMENTS
■
The authors want to thank Katharina Bauer for biological
testing and Ida D’Orazio for the synthesis of compounds 57
and 80.
REFERENCES
■
(1) Dalrymple, S. A. p38 mitogen activated protein kinase as a
therapeutic target for Alzheimer’s disease. J. Mol. Neurosci. 2002, 19,
295−299.
(2) den Broeder, A.; van de Putte, L.; Rau, R.; Schattenkirchner, M.;
Van Riel, P.; Sander, O.; Binder, C.; Fenner, H.; Bankmann, Y.;
Velagapudi, R.; Kempeni, J.; Kupper, H. A single dose, placebo
controlled study of the fully human anti-tumor necrosis factor-alpha
antibody adalimumab (D2E7) in patients with rheumatoid arthritis. J.
Rheumatol. 2002, 29, 2288−2298.
(3) Feldmann, M.; Brennan, F. M.; Maini, R. N. Rheumatoid arthritis.
Cell 1996, 85, 307−310.
(4) Mikuls, T. R.; Moreland, L. W. TNF blockade in the treatment of
rheumatoid arthritis: infliximab versus etanercept. Expert Opin.
Pharmacother. 2001, 2, 75−84.
(5) Cohen, S.; Hurd, E.; Cush, J.; Schiff, M.; Weinblatt, M. E.;
Moreland, L. W.; Kremer, J.; Bear, M. B.; Rich, W. J.; McCabe, D.
Treatment of rheumatoid arthritis with anakinra, a recombinant
human interleukin-1 receptor antagonist, in combination with
methotrexate: Results of a twenty-four-week, multicenter, randomized,
double-blind, placebo-controlled trial. Arthritis Rheum. 2002, 46, 614−
624.
(6) Lee, J. C.; Laydon, J. T.; McDonnell, P. C.; Gallagher, T. F.;
Kumar, S.; Green, D.; McNulty, D.; Blumenthal, M. J.; Heys, J. R.;
Landvatter, S. W. A protein kinase involved in the regulation of
inflammatory cytokine biosynthesis. Nature 1994, 372, 739−746.
(7) Karcher, S. C.; Laufer, S. A. Successful structure-based design of
recent p38 MAP kinase inhibitors. Curr. Top. Med. Chem. (Sharjah,
United Arab Emirates) 2009, 9, 655−676.
(8) Lee, M. R.; Dominguez, C. MAP kinase p38 inhibitors: Clinical
results and an intimate look at their interactions with p38alpha protein.
Curr. Med. Chem. 2005, 12, 2979−2994.
Diffraction data were collected at the PX10SA beamline of the Swiss
Light Source (PSI, Villingen, Switzerland) to a resolution of 2.4 Å
using wavelengths close to 1 Å. Images were processed with XDS and
scaled using XSCALE.44
Structure Determination and Refinement of p38α with
Compound 45. The p38α−inhibitor complex structure was obtained
by molecular replacement with PHASER,45 using a related p38α MAP
kinase structure (Protein Data Bank entry 3QUE24) as a search model.
Crystallographic refinement was performed with CNS46 and
Refmac5.47 Model building and real space refinement was done
using the program COOT.48 The topology and library files for the
ligand were generated with the Dundee PRODRG server.49 The
refined structure was validated with PROCHECK.50 Detailed data,
refinement, and geometry statistics are provided in Table 4. The
graphics were obtained using PyMOL 1.5.0.4.
(9) Liu, Y.; Gray, N. S. Rational design of inhibitors that bind to
inactive kinase conformations. Nat. Chem. Biol. 2006, 2, 358−364.
(10) Pettus, L. H.; Wurz, R. P. Small molecule p38 MAP kinase
inhibitors for the treatment of inflammatory diseases: novel structures
and developments during 2006−2008. Curr. Top. Med. Chem. 2008, 8,
1452−1467.
(11) Kumar, S.; Jiang, M. S.; Adams, J. L.; Lee, J. C.
Pyridinylimidazole compound SB 203580 inhibits the activity but
not the activation of p38 mitogen-activated protein kinase. Biochem.
Biophys. Res. Commun. 1999, 263, 825−831.
(12) Kumar, S.; Boehm, J.; Lee, J. C. p38 MAP kinases: Key
signalling molecules as therapeutic targets for inflammatory diseases.
Nat. Rev. Drug Discovery 2003, 2, 717−726.
(13) Moran, N. p38 kinase inhibitor approved for idiopathic
pulmonary fibrosis. Nat. Biotechnol. 2011, 29, 301.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental and analytical data, including HPLC purity and
HRMS data of compounds 17−88 and selectivity profile of
compound 45. This material is available free of charge via the
Internet at http://pubs.acs.org.
(14) Xing, L.; Shieh, H. S.; Selness, S. R.; Devraj, R. V.; Walker, J. K.;
Devadas, B.; Hope, H. R.; Compton, R. P.; Schindler, J. F.; Hirsch, J.
L.; Benson, A. G.; Kurumbail, R. G.; Stegeman, R. A.; Williams, J. M.;
Broadus, R. M.; Walden, Z.; Monahan, J. B. Structural bioinformatics-
based prediction of exceptional selectivity of p38 MAP kinase inhibitor
PH-797804. Biochemistry 2009, 48, 6402−6411.
Accession Codes
A new structure of the human p38α MAP kinase in complex
with compound 45 is deposited under the accession code
3UVP in the Protein Data Bank.
(15) Pharmacia & Upjohn Co.LLC. Crystalline forms of 3-[5-chloro-
4-[(2,4-difluorobenzyl) oxy]-6-oxopyrimidin-1(6H)-YL]-N-(2-hydrox-
7872
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
yethyl)-4-methylbenzamide. International Patent WO2006040649,
2006.
(16) Pharmacia Corp. Diaryl substituted pyridinones. U.S. Patent
US2006211694, 2006.
(17) Pharmacia Corp. Substituted pyrimidinones. U.S. Patent
US7795271, 2010.
the traditional pharmacophore approach of type I inhibition. Expert
Opin. Drug Discovery 2012, 3, 1409−1425.
(33) Backes, A. C.; Zech, B.; Felber, B.; Klebl, B.; Mueller, G. Small-
molecule inhibitors binding to protein kinase. Part II: the novel
pharmacophore approach of type II and type III inhibition. Expert
Opin. Drug Discovery 2008, 3, 1427−1449.
(34) Cogan, D. A.; Aungst, R.; Breinlinger, E. C.; Fadra, T.;
Goldberg, D. R.; Hao, M. H.; Kroe, R.; Moss, N.; Pargellis, C.; Qian,
K. C.; Swinamer, A. D. Structure-based design and subsequent
optimization of 2-tolyl-(1,2,3-triazol-1-yl-4-carboxamide) inhibitors of
p38 MAP kinase. Bioorg. Med. Chem. Lett. 2008, 18, 3251−3255.
(35) Wrobleski, S. T.; Lin, S.; Hynes, J., Jr.; Wu, H.; Pitt, S.; Shen, D.
R.; Zhang, R.; Gillooly, K. M.; Shuster, D. J.; McIntyre, K. W.;
Doweyko, A. M.; Kish, K. F.; Tredup, J. A.; Duke, G. J.; Sack, J. S.;
McKinnon, M.; Dodd, J.; Barrish, J. C.; Schieven, G. L.; Leftheris, K.
Synthesis and SAR of new pyrrolo[2,1-f][1,2,4]triazines as potent p38
alpha MAP kinase inhibitors. Bioorg. Med. Chem. Lett. 2008, 18, 2739−
2744.
(18) Norman, P. BMS-582949: crystalline form of a p38alpha
inhibitor? WO2008079857. Expert Opin. Ther. Pat. 2009, 19, 1165−
1168.
(19) Malley, M. F. (Bristol-Myers Squibb Company, USA)
Crystalline forms of an aryl-substituted pyrazoleamide. International
Patent WO 2008079857. 2008.
(20) Liu, C.; Lin, J.; Wrobleski, S. T.; Lin, S.; Hynes, J.; Wu, H.;
Dyckman, A. J.; Li, T.; Wityak, J.; Gillooly, K. M.; Pitt, S.; Shen, D. R.;
Zhang, R. F.; McIntyre, K. W.; Salter-Cid, L.; Shuster, D. J.; Zhang, H.;
Marathe, P. H.; Doweyko, A. M.; Sack, J. S.; Kiefer, S. E.; Kish, K. F.;
Newitt, J. A.; McKinnon, M.; Dodd, J. H.; Barrish, J. C.; Schieven, G.
L.; Leftheris, K. Discovery of 4-(5-(cyclopropylcarbamoyl)-2-methyl-
phenylamino)-5-methyl-N-propylpyrrolo[1,2-f ][1,2,4]triazine-6-car-
boxamide (BMS-582949), a clinical p38α MAP kinase inhibitor for the
treatment of inflammatory diseases. J. Med. Chem. 2010, 53, 6629−
6639.
(36) www.proqinase.com. 2012.
(37) Scios Inc. Indole-type inhibitors of p38 kinase. U.S. Patent
US6890938. 2005.
(38) Simard, J. R.; Getlik, M.; Grutter, C.; Schneider, R.; Wulfert, S.;
Rauh, D. Fluorophore labeling of the glycine-rich loop as a method of
identifying inhibitors that bind to active and inactive kinase
conformations. J. Am. Chem. Soc. 2010, 132, 4152−4160.
(39) Boehringer Ingelheim Pharma. Methods of treating COPD and
pulmonary hypertension. International Patent WO200501862. 2005.
(40) Ahn, Y. M.; Clare, M.; Ensinger, C. L.; Hood, M. M.; Lord, J.
W.; Lu, W. P.; Miller, D. F.; Patt, W. C.; Smith, B. D.; Vogeti, L.;
Kaufman, M. D.; Petillo, P. A.; Wise, S. C.; Abendroth, J.; Chun, L.;
Clark, R.; Feese, M.; Kim, H.; Stewart, L.; Flynn, D. L. Switch control
pocket inhibitors of p38-MAP kinase. Durable type II inhibitors that
do not require binding into the canonical ATP hinge region. Bioorg.
Med. Chem. Lett. 2010, 20, 5793−5798.
(41) Getlik, M.; Grutter, C.; Simard, J. R.; Nguyen, H. D.; Robubi,
A.; Aust, B.; van Otterlo, W. A.; Rauh, D. Structure-based design,
synthesis and biological evaluation of N-pyrazole, N′-thiazole urea
inhibitors of MAP kinase p38alpha. Eur. J. Med. Chem. 2012, 48, 1−15.
(42) Simard, J. R.; Getlik, M.; Grutter, C.; Pawar, V.; Wulfert, S.;
Rabiller, M.; Rauh, D. Development of a fluorescent-tagged kinase
assay system for the detection and characterization of allosteric kinase
inhibitors. J. Am. Chem. Soc. 2009, 131, 13286−13296.
(43) Bukhtiyarova, M.; Northrop, K.; Chai, X.; Casper, D.; Karpusas,
M.; Springman, E. Improved expression, purification, and crystal-
lization of p38alpha MAP kinase. Protein Expression Purif. 2004, 37,
154−161.
(44) Kabsch, W. Automatic processing of rotation diffraction data
from crystals of initially unknown symmetry and cell constants. J. Appl.
Crystallogr. 1993, 26, 795−800.
(45) Read, R. J. Pushing the boundaries of molecular replacement
with maximum likelihood. Acta Crystallogr., Sect. D: Biol. Crystallogr.
2001, 57, 1373−1382.
(46) Brunger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.;
Gros, P.; Grosse-Kunstleve, R. W.; Jiang, J. S.; Kuszewski, J.; Nilges,
M.; Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.; Warren, G. L.
Crystallography & NMR system: A new software suite for macro-
molecular structure determination. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 1998, 54, 905−921.
(21) Lee, J. C.; Kassis, S.; Kumar, S.; Badger, A.; Adams, J. L. p38
mitogen-activated protein kinase inhibitors–mechanisms and ther-
apeutic potentials. Pharmacol. Ther. 1999, 82, 389−397.
(22) Fischer, S.; Koeberle, S. C.; Laufer, S. A. p38alpha mitogen-
activated protein kinase inhibitors, a patent review. Expert Opin. Ther.
Pat. 2011, 21, 1843−1866.
(23) Laufer, S. A.; Ahrens, G. M.; Karcher, S. C.; Hering, J. S.; Niess,
R. Design, synthesis, and biological evaluation of phenylamino-
substituted 6,11-dihydro-dibenzo[b,e]oxepin-11-ones and dibenzo-
[a,d]cycloheptan-5-ones: Novel p38 MAP kinase inhibitors. J. Med.
Chem. 2006, 49, 7912−7915.
(24) Koeberle, S. C.; Romir, J.; Fischer, S.; Koeberle, A.; Schattel, V.;
Albrecht, W.; Grutter, C.; Werz, O.; Rauh, D.; Stehle, T.; Laufer, S. A.
Skepinone-L is a selective p38 mitogen-activated protein kinase
inhibitor. Nat. Chem. Biol. 2011, 8, 141−143.
(25) Murineddu, G.; Ruiu, S.; Loriga, G.; Manca, I.; Lazzari, P.; Reali,
R.; Pani, L.; Toma, L.; Pinna, G. A. Tricyclic pyrazoles. 3. Synthesis,
biological evaluation, and molecular modeling of analogues of the
cannabinoid antagonist 8-chloro-1-(2′,4′-dichlorophenyl)-N-piperidin-
1-yl-1,4,5,6-tetrahydrobenzo[6,7]cyclohepta[1,2-c]pyrazole-3-carboxa-
mide. J. Med. Chem. 2005, 48, 7351−7362.
(26) Engelhardt, E. L.; Zell, H. C.; Saari, W. S.; Christy, M. E.;
Colton, C. D.; Stone, C. A.; Stavorski, J. M.; Wenger, H. C.; Ludden,
C. T. Structure−activity relationships in the cyproheptadine series. J.
Med. Chem. 1965, 8, 829−835.
(27) Kurokawa, M.; Sato, F.; Masuda, Y.; Yoshida, T.; Ochi, Y.;
Zushi, K.; Fujiwara, I.; Naruto, S.; Uno, H.; Matsumoto, J. Synthesis
and biological activity of 11-[4-(cinnamyl)-1-piperazinyl]- 6,11-
dihydrodibenz[b,e]oxepin derivatives, potential agents for the treat-
ment of cerebrovascular disorders. Chem. Pharm. Bull. (Tokyo) 1991,
39, 2564−2573.
(28) Schrodinger Suite 2008 Induced Fit Docking protocol. [Glide
̈
version 5.0] Schrodinger, LLC, New York, NY, 2005.
̈
(29) Jensen, T. A.; Liang, X.; Tanner, D.; Skjaerbaek, N. Rapid and
efficient microwave-assisted synthesis of aryl aminobenzophenones
using Pd-catalyzed amination. J. Org. Chem. 2004, 69, 4936−4947.
(30) Goettert, M.; Graeser, R.; Laufer, S. A. Optimization of a
nonradioactive immunosorbent assay for p38alpha mitogen-activated
protein kinase activity. Anal. Biochem. 2010, 406, 233−234.
(31) Angell, R.; Aston, N. M.; Bamborough, P.; Buckton, J. B.;
Cockerill, S.; deBoeck, S. J.; Edwards, C. D.; Holmes, D. S.; Jones, K.
L.; Laine, D. I.; Patel, S.; Smee, P. A.; Smith, K. J.; Somers, D. O.;
Walker, A. L. Biphenyl amide p38 kinase inhibitors 3: Improvement of
cellular and in vivo activity. Bioorg. Med. Chem. Lett. 2008, 18, 4428−
4432.
(47) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Refinement of
macromolecular structures by the maximum-likelihood method. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 1997, 53, 240−255.
(48) Emsley, P.; Cowtan, K. Coot: Model-building tools for
molecular graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004,
60, 2126−2132.
(49) Schuttelkopf, A. W.; van Aalten, D. M. PRODRG: A tool for
high-throughput crystallography of protein-ligand complexes. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2004, 60, 1355−1363.
(32) Backes, A. C.; Zech, B.; Felber, B.; Klebl, B.; Mueller, G. Small-
molecule inhibitors binding to protein kinases. Part I: exceptions from
7873
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874

Journal of Medicinal Chemistry
Article
(50) Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.; Thornton, J.
M. PROCHECK: A program to check the stereochemical quality of
protein structures. J. Appl. Crystallogr. 1993, 26, 283−291.
7874
dx.doi.org/10.1021/jm300951u | J. Med. Chem. 2012, 55, 7862−7874