A frozen analogue approach to aminopyridinylimidazoles leading to novel and promising p38 MAP kinase inhibitors

Article abstract of DOI:10.1021/jm300852w

In this study we report the design, synthesis, and biological evaluation of constrained aminopyridinylimidazoles as p38α MAP kinase inhibitors. The frozen analogue approach focused on the pyridinyl unit, using purine bioisosteres as constrained structure

Full text of DOI:10.1021/jm300852w

Article
pubs.acs.org/jmc
A Frozen Analogue Approach to Aminopyridinylimidazoles Leading
to Novel and Promising p38 MAP Kinase Inhibitors
Roland Selig,† Marcia Goettert,† Verena Schattel,† Dieter Schollmeyer,‡ Wolfgang Albrecht,§
and Stefan Laufer*,†
†Department of Medicinal Chemistry, University of Tuebingen, Auf der Morgenstelle 8, 72076 Tuebingen, Germany
‡Department of Organic Chemistry, University of Mainz, Duesbergweg 10-14, 55099 Mainz, Germany
§c-a-i-r biosciences GmbH, Paul-Ehrlich-Strasse 15, 72076 Tuebingen, Germany
S
* Supporting Information
ABSTRACT: In this study we report the design, synthesis, and
biological evaluation of constrained aminopyridinylimidazoles as
p38α MAP kinase inhibitors. The frozen analogue approach focused
on the pyridinyl unit, using purine bioisosteres as constrained
structure analogues. The identification of the most potent bioisostere
was followed by a further derivatization to address hydrophobic
region II. In combination with C-2 modifications of the imidazole
core, we were able to design highly active inhibitors on the p38α
MAP kinase. The inhibitor design presented herein represents a
promising and highly efficient advancement of recent stages of
development in this class of p38 MAP kinase inhibitors. In
combination with the highly flexible synthetic strategy, directions
toward further investigations of complex C-5 modifications of
diarylimidazoles are indicated.
effect of N-substitutions on inhibitor potency, showing that N-
substituted pyridinylimidazoles bind to p38 MAP kinase with a
generally lower affinity than their N-unsubstituted counter-
parts.4,6−8 In a further refinement, alkylated or arylated amines
were introduced at the C-2 position of the pyridine moiety, a
modification that improved potency and selectivity by
formation of an additional hydrogen bond to the hinge region
and addressing hydrophobic region II (Figure 1).9 This kind of
modification in combination with the C-2 modifications of the
imidazole core lead to highly selective and potent inhibitors
with IC50 values in the low nanomolar range.
The binding of the ligand to the receptor results in an energy
penalty due to the loss of conformational energy that is
detrimental to the strength of the binding. Conformational
restriction is a commonly used principle to avoid these adverse
effects. Reduction in conformational flexibility can be realized
by steric hindrance or restriction, which minimizes the
entropy−enthalpy compensation by minimizing the loss of
conformational entropy11,12 and therefore enhance potency as
well as selectivity of the corresponding ligands. As investigated
by Veber et al., the characteristics of the ligands are markedly
improved by freezing rotatable bonds. In this study a
correlation between the bioavailability of a compound and
the number of rotatable bonds was shown, irrespective of the
INTRODUCTION
■
The inhibition of interleukin 1β (IL-1β) production by the anti-
inflammatory compound SKF860021 (Figure 1) was first
reported in 1988,1 but it was not until 1994 that the molecular
target of this class of pyridinylimidazole compounds was
identified as p38α mitogen activated protein kinase (MAPK), a
serine/threonine kinase. Because of the central role of p38
MAPK in inflammatory bowel disease and rheumatoid arthritis,
the inhibition of p38 MAPK became a promising therapeutic
strategy. The further development of p38 MAPK inhibitors
focused on derivatizations of the core structure of SKF86002,
leading to the early lead compound SB2035802 (Figure 1) with
an additional interaction site, the C-2 aryl moiety forming a
π−π stacking to Tyr35.3 The development of this kind of p38
MAP kinase inhibitors followed a straightforward inhibitor
design. The vicinal pyridine 4-fluorophenyl system within this
pharmacophore proved to be one of the most efficient and
selective systems for binding to p38 MAP kinase, and this
structure has been maintained in many p38 inhibitors modeled
on SB 203580.4
The pyridine nitrogen accepts a hydrogen bond from the
backbone of Met109, and the fluorophenyl moiety confers
selectivity for MAPK over other PKs because of its interaction
with hydrophobic region I (Figure 1).5
Further derivatizations were focused on the N1-substituted
imidazoles, mainly to reduce interactions with CYP450
enzymes.4 Boehm and Gallagher reported the detrimental
Received: June 17, 2012
Published: September 6, 2012
© 2012 American Chemical Society
8429
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
Figure 1. Structures of pyridinyl imidazole compound SKF86002 and derivatives SB203580 (left) and ML3403 (right), showing binding mode of
SB203580 and sites of derivatization of SB203580 (R1= various residues, aryl, SMe, H, alkinyl; R2= hydrophobic residues; R3 = H, alkyl) and
ML340310 (R1 = SMe; R2 = phenyl; R3 = Me).
Figure 2. Rigidization concept based on ML3403.
molecular weight. The resulting findings of this study showed
that even the Lipinsky rule in regard to the molecular weight
might be circumvented using a frozen analogue approach.13
Taken in this context, we proposed to further develop the
class of aminopyridinylimidazoles using a rigidization strategy
and to focus on the most effective protein−ligand interaction
site, the C-5 scaffold. A highly effective, literature-known
aminopyridine residue proved to be the methylphenylamine
moiety of ML3403 (Figure 1), which served as a starting point
for our investigations. Maintaining the highly effective donor−
acceptor system of the aminopyridinyl system, we investigated
purine bioisosteres as new C-5 rigidized aminopyridinyl
scaffolds of the imidazole core with implemented aromatization
(Figure 2). Among the various C-2 modifications at the
imidazole core, we chose, in contrast to ML3403, phenylation
of the C-2 residue in order to maintain the π−π stacking
interaction with Tyr35.
CHEMISTRY
■
We followed a straightforward synthetic procedure involving a
convergent strategy (Scheme 1) that was based on C−C cross-
coupling reactions, as recently published by our group.14 This
high yielding synthesis was optimized by ligand screening and
technical modifications with regard to mechanistic details.
The challenge of this cross-coupling reaction was to combine
activated (purine and pyrrolopyrimidine) and deactivated
substrates (azaindoles) as cross-coupling partners in a single
robust process. The deactivated chlorinated azaindoles were
especially sensitive to the destannylation process as the
dominant side reaction, lowering the yields and reaction rates
dramatically. In our investigations of the detrimental side
reactions and reaction rates, changing the ligand sphere from
the initially used PPh3 to X-Phos proved to be the most
effective modification. The high efficiency of the X-Phos ligand
system is proposed to arise from the monocoordinated Pd
species15 that facilitates both the oxidative addition and the
transmetalation step. The mechanism of the Stille coupling with
8430
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthesis of the Tested Compounds
Scheme 3. Representative Cross-Coupling Reaction of 6 and
4-(4-Fluorophenyl)-1-(methoxymethyl)-2-phenyl-5-
(tributylstannyl)-1H-imidazole18 (7) in 2 mL of DMF at 83
°C under Argon Atmosphere
a
X = CH, N; pg = protecting group; R = H, cyclohexyl, phenyl, benzyl,
annelated phenyl.
PPh3 was substrate specific with the electron poor heterocycles
6-chloro-9-(tetrahydro-2H-pyran-2-yl)-9H-purine16 (4) and 4-
chloro-7-(methoxymethyl)-7H-pyrrolo[2,3-d]pyrimidine14 (5),
following the established rate determining step, the trans-
metalation. In contrast, the rate determining step of the Stille
coupling with 4-chloro-1-(phenylsulfonyl)-1H-pyrrolo[2,3-b]-
pyridine17 6 was determined to be the oxidative addition
(Scheme 2), when using PPh3 as the ligand system.14
Scheme 2. Effect of the Substrate on the Stille Cross-
Coupling and the Rate Determining Steps
Figure 3. Effect of the concentration of the stannylimidazole on the
turnover numbers (TON) in the cross-coupling reaction as a function
of time at 83 °C in 3 mL of DMF, quantified by HPLC.
chlorinated substrates would be more advantageous than the
bromo or iodo species for the cross-coupling process. The
cross-coupling reaction to obtain 7 was performed with 6 and
the corresponding stannylimidazole 3 (Scheme 3).
Since the mechanistic course of the cross-coupling reaction is
influenced by the nature of the ligands, factors that stabilize the
essential intermediates of the Pd species ought to further
increase the reaction rates. We therefore examined the
stabilizing effects of the solvents.
For an optimization of the cross-coupling reaction with PPh3,
an accelerating effect of the solvent would be expected to assist
the rate determining oxidative addition.
The rate determining step with X-Phos was shown to be the
transmetalation. Coordinating solvents should also play a
beneficial role in this step because of their capability of
stabilizing the three coordinated intermediates. The solvents
used in our study ranged from weakly coordinating to strongly
coordinating solvents, ranked as follows: toluene, THF,
dioxane, DMSO, NMP, DMF.
Using quantitative HPLC, we investigated the effects of the
solvent on the turnover frequencies of the two different catalyst
systems at 76 °C (Figure 4). As representative substrates for
the cross-coupling reaction, we chose the chlorinated phenyl-
sulfonyl protected 7-azaindole 6 and the stannylated diary-
limidazole 7 as substrates (Figure 4). Toluene and dioxane
were the most effective solvents in the cross-coupling reaction
with PdCl2(PPh3)2 as precatalyst. DMSO showed a moderate
turnover frequency, while the reaction rates with DMF and
NMP were very low (Figure 4).
In the present study, we have extended our chemical
investigations of this synthetic procedure, based on a solvent
screening approach with the intention of further optimization
of the existing synthetic procedure.
First, we endeavored to clarify the question of the rate
determining step in the Stille cross-coupling when using 2-
dicyclohexylphosphino-2′,4′,6′-triisopropylbiphenyl (X-Phos)
as the ligand system and 6 as the substrate for oxidative
addition (Scheme 2). A change in the mechanism with X-Phos
is likely. In our published study we showed that the oxidative
addition was the rate determining step in the case of
triphenylphosphine (PPh3) as the ligand system. With X-
Phos, we presumed that the transmetalation had become the
rate determining step. Our measurements of the time-
dependent turnover numbers (Scheme 3) revealed a clear
dependency on the concentration of the stannylated compound
3 (Figure 3). The direct dependence on the stannyl
concentration indicated that the transmetalation was the rate
determining step. This inversion of the rate determining step to
favor the transmetalation reaction with 6 indicated that the
Weakly coordinating and moderately coordinating solvents
were favored, which is in contrast to the expected results. The
8431
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
have prevented effective coordination of Pd to the weakly
electrophilic center of 6. The reduced coordination capacity of
toluene and dioxane seems to be crucial in forming the
oxidative addition complex in the case of the deactivated
halogen substrates, such as 6, possibly because of the lower
substrate−Pd affinity which is further reduced by the
competing solvent−Pd coordination.
Further investigations focused on the highly efficient X-Phos
ligand system (Figure 5). The solvent screening of this system
was analogous to that of PPh3.
Figure 4. Effect of the solvent on the turnover frequencies in the cross-
coupling reaction performed at 76 °C with PdCl2(PPh3)2 as precatalyst
(Table 1, entries 1−5).
Table 1. Optimization of the Cross-Coupling Reaction:
Comparison of the Effects of Pd Ligand, Solvent, Reaction
Temperature, and Reaction Time
Figure 5. Effect of the solvent on the turnover frequencies in the cross-
coupling reaction performed at 76 °C with a premilled mixture of
Pd(OAc)2 and X-Phos (1/3) as precatalyst (Table 1, entries 6−10).
yield
a
b
entry
catalyst
conditions
t [min]
[%]
1
2
3
4
5
6
PdCl2(PPh3)2, 17 mol %
PdCl2(PPh3)2, 16 mol %
PdCl2(PPh3)2, 18 mol %
PdCl2(PPh3)2, 14 mol %
PdCl2(PPh3)2, 16 mol %
DMF/76 °C
NMP/76 °C
DMSO/76 °C
toluene/76 °C
dioxane/76 °C
DMF/76 °C
30
30
30
30
30
30
3
A different reactivity profile was obtained with X-Phos that
was at odds with the predicted order of the solvent
coordination capacity. Moderately coordinating solvents were
disfavored in the transmetalation step with X-Phos as ligand
sphere. In contrast to PPh3 (Figure 4), strongly coordinated
DMF proved to be the most effective solvent with X-Phos
(Figure 5).
Compared to dioxane, NMP and toluene also strongly
accelerated the reaction rates, whereas DMSO afforded lower
turnover frequencies. Coordinating solvents showed a positive
influence. The most attractive solvent in our recently published
study was dioxane, which proved to be also the most efficient
solvent with PPh3 in the present study. However, using dioxane
with the more efficient X-Phos ligand system was disadvanta-
geous, giving a 4-fold slower reaction rate than DMF (Figure
5).
To further investigate the accelerating solvent effect of DMF
with X-Phos, we examined the effect of temperature on the
turnover frequencies (Figure 6). The optimum turnover
frequency was obtained at 102 °C and was in the range of
our earlier study with microwave irradiation.14 Thus, conven-
tional heating with DMF as solvent led to highly efficient
conversions at low temperatures.
2
4
10
13
93
Pd(OAc)2/X-Phos, 10
mol %
7
8
Pd(OAc)2/X-Phos, 10
mol %
NMP/76 °C
30
30
78
8
Pd(OAc)2/X-Phos, 10
mol %
DMSO/76 °C
9
Pd(OAc)2/X-Phos 10 mol % toluene/76 °C
Pd(OAc)2/X-Phos, 10
mol %
30
30
47
22
10
dioxane/76 °C
DMF/76 °C
DMF/95 °C
DMF/102 °C
DMF/76 °C
DMF/76 °C
11
12
13
14
15
Pd(OAc)2/X-Phos, 12
mol %
10
30
30
20
67
15
51
92
24
2
Pd(OAc)2/X-Phos, 11
mol %
Pd(OAc)2/X-Phos, 10
mol %
Pd(OAc)2/X-Phos, 12
mol %
Pd(OAc)2/X-Phos, 12
mol %
a
Conditions: 1 equiv of 6, 2 equiv of 3 in 3 mL of solvent for indicated
reaction time. Quantified by HPLC.
b
In conclusion, the cross-coupling procedure was further
optimized by solvent assisting effects. The high reaction rates
showed the efficiency of the catalytic system even at 102 °C,
without the need for microwave irradiation. The change of the
strong coordination of DMF leading to the PPh3-Pd-DMF/
[DMF-Pd-Cl]− species, as assumed by Ahlquist et al.,19 may
8432
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
HClaq in refluxing MeOH (Scheme 4) to remove both MOM
groups.23
BIOLOGICAL EVALUATIONS
■
On the basis of our concept to investigate purine bioisosteres as
new rigidized C-5 scaffolds of the aminopyridinyl moieties, we
developed a strategy based on five stages (Figure 7). Stage I
Figure 6. Effect of temperature on the turnover frequencies in the
cross-coupling reaction in the presence of X-Phos as Pd ligand (Table
1, entries 11−13).
solvent from dioxane to DMF allowed an expanded temper-
ature range with the option of even higher reaction rates. With
these reaction rates, the catalyst loadings can be significantly
reduced to low levels.
The choice of the protecting groups was applied to the final
target compounds (Scheme 4). The investigation of N-
Figure 7. Strategic concept for investigation of purine bioisosteres.
investigations focused on the identification of the most potent
purine bioisostere. In stage II, we investigated the influence of
N-substituents on the imidazole core. The objective of stage III
was to address the hydrophobic region II of MAPK by further
lipophilic modifications of the most effective purine bioisostere
in the inhibitors. The investigations of stage IV examined the
impact of further enhancing or diminishing the rigidization of
the substituted purine scaffold. The final stage V involved
improving the potency of the evaluated inhibitor with
literature-known modifications of the 2-phenyl residue at the
imidazole core.
Scheme 4. Preparation of 10 and 12: Deprotection
Procedures
Our incipient effort at inhibitor design at stage I was to
develop a potent isosteric scaffold of the aminopyridinyl
moiety. The 7-azaindole is the most obvious rigidized structural
analogue of the aminopyridinyl unit, but we also examined a
series of bioisosteres with purine and pyrrolopyrimidine
substituents as potential candidates (14a and 14b) (Figure 8).
To directly compare biological activity, we chose a fixed core
structure; in the case of C-5 derivatizations, we examined the
C-2 phenylation of 4-(4-fluorophenyl)-2-phenylimidazole
(Figure 8, structure 14/15).
substitution led us to select an orthogonal strategy with 4-
chloro-1-tosyl-1H-pyrrolo[2,3-b]pyridine20 (8) and the me-
thoxymethyl (MOM) protected imidazole core. The detosyla-
tion was accomplished with Cs2CO321,22 in refluxing methanol,
leading to the MOM protected target compound 4-(4-(4-
fluorophenyl)-1-(methoxymethyl)-2-phenyl-1H-imidazol-5-yl)-
1H-pyrrolo[2,3-b]pyridine (10) (Scheme 4).
In this study the most potent structures in regard to their
biological activities were shown to be the N-unsubstituted
analogues of 12. For their synthesis we chose a parallel
deprotection procedure with methoxmethyl as the dual
protecting group in compound 13.6 The deprotection
procedure followed established methods, using concentrated
Figure 8. C-5 modifications of the imidazole core.
8433
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
With IC50 values of 740 nM (14a), 2600 nM (14b), and 108
nM (14c), the 7-azaindole substituent was identified as the
most potent C-5 scaffold (Table 2 and Figure 10). Compared
Table 2. Inhibition of p38 MAPK
activity coefficient,
IC50(SB203580)/IC50(compd)
a
compd
IC50 [μM]
14a
14b
14c
14d
15a
15c
0.740 0.1169
2.600 0.0862
0.108 0.0012
0.154 0.0092
2.989 0.0582
0.451 0.0589
0.10
0.03
0.47
0.33
0.01
0.18
a
Mean SEM of three experiments. Compounds tested according to
Laufer et al.24
Figure 10. Lipophilic substitution of 7-azaindole to address the
hydrophobic region II.
to the reference compound 4-(4-(4-fluorophenyl)-2-phenyl-
1H-imidazol-5-yl)pyridine (14d), a 1.4-fold increase in potency
with an activity coefficient of 0.47 (IC50(SB203580 (2))/
IC50(compound)) was assigned to 14c.
Maintaining the most potent C-5 scaffold, the 7-azaindole,
we investigated in stage II the impact of the N1-derivatizations,
widely used optimization sites in literature compounds and
mainly used to reduce CYP450 interactions.7 We assumed a
negative effect of these derivatizations, due to disadvantageous
conformational effects of the azaindole moiety, which is flipping
out of the biological relevant conformation by the repulsion of
the N-residues (Figure 9). The flexibility of the known
respect to the reference compound 14d (Table 2) (compared
by activity coefficients).
Table 3. Inhibition of p38 MAPK by Substituents That Affect
the Interactions with Hydrophobic Region II
activity coefficient,
IC50(SB203580)/IC50(compd)
a
compd
IC50 [μM]
14e
14f
0.153 0.0398
0.016 0.0038
0.345 0.0319
0.015 0.0003
0.55
2.38
0.07
2.07
14g
14h
a
Mean SEM of three experiments. Compounds tested according to
Laufer et al.24
To enhance the entropic effect, we dearomatized the phenyl
moiety to the more bulky cyclohexyl analogue 14h (Figure 10)
in order to expand the lipophilic surface. Nevertheless, the
phenyl residue of 14f (Figure 10) proved to be the more
efficient residue to address the hydophobic region II with an
activity coefficient of 2.38, whereas the activity coefficient of
14h was shown to be slightly lowered with 2.07 (Table 3).
The strategy of stage IV was to examine the impact of
rigidization on the 2-phenylated azaindole scaffold. With α-
carboline the maximum rigidized structure analogue (14e)
(Figure 10) to address hydrophobic region II was synthesized.
With an IC50 of 153 nM and an activity coefficient of 0.55 the
carboline 14e showed only a slightly increased activity profile
(Table 3) compared to the unsubstituted azaindole derivative
14c (Table 2), despite the extended possible hydrophobic
interaction side provided by the annelated phenyl residue.
Two disadvantageous synergistic effects were assumed to be
responsible: the lack of flexibility and the steric demand of the
carboline. Because of the annelated phenyl moiety, the steric
demand of the azaindole is negatively expanded in the direction
to the imidazole core, resulting in a putative increased energy
penalty to the biologically relevant conformation. Docking
studies carried out to evaluate these assumptions confirmed a
disturbed π−π stacking with Tyr35 (Figure 11), due to the
steric repulsion of the carboline plane from the imidazole plane,
with the consequence of increased interaction distances. The
obvious lack of flexibility of the phenyl moiety with respect to
the hydrophobic interaction with the hydrophobic region II is a
further detrimental effect of this rigidization. The loss of the
Figure 9. Conformational effects of N1 substituents.
aminopyridinylimidazoles provides a better conformational
tolerance, due to the possible independent torsion oppor-
tunities of the free rotatable C−N−C bonds. This rigidization,
leading to the 7-azaindole, results in the repulsion of the entire
C-5 moiety, effectively disturbing the hinge region interaction.
The decreased conformational tolerance of the new rigidized
structure analogues is reflected in their loss of biological
activities, with a 10- and nearly 3-fold decrease in potency,
respectively, of compounds 15a and 15c compared to their
unsubstituted analogues 14a and 14c (Table 2).
In stage III, the further development of the 7-azaindole
moiety focused on addressing interactions with the hydro-
phobic region II (Figure 10). Among the wide range of
accomplished modifications in recent years, phenyl substituents
such as phenylethylamine9 have been among the most
promising residues addressing this region. Therefore, we
expanded the azaindole scaffold to include the 2-phenylated
system, the rigidized analogue to the aminopyridine moiety of
ML3403 (Figure 2). With an IC50 of 16 nM (Table 3) the
compound 14f showed a 7-fold improved activity profile with
8434
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
Figure 11. Docking results for 14e in the p38 MAP kinase. X-ray
Figure 12. Docking results for 14f in the p38 MAP kinase. X-ray
structure 1YWR20 is from the PDB database.
structure 1YWR25 is from the PDB database.
The 2 phenylated azaindol system differs markedly from this
orientation as shown in Figure 13.
C−N−C torsion possibilities and the predetermined prejudicial
orientation of the annelated phenyl residue lead to a reduced
capability for ideal interactions with hydrophobic region II.
The disadvantageous elements of the carboline scaffold may
be avoided by using the aforementioned 2-phenyl substituted
azaindole derivative 14f (Figure 10). The essential flexibility is
given, and a possible repulsion with the imidazole core is
circumvented by the sterically neutral orientation of the phenyl
substituent.
Docking experiments with 14f indicated advantageous
torsion angles for the diarylimidazole and the azaindole plane
for interactions with Tyr35 (Figure 12). The C-2 (azaindole)
phenyl moiety and the azaindole are slightly twisted in relation
to one another, and the orientation of the phenyl residue,
addressing hydrophobic region II, is markedly different from
that of the carboline substituted derivative 14e (Figures 11 and
12), leading to an improved hydrophobic interaction.
A further loss of rigidization by the benzylated azaindole 14g
(Figure 10) led to a dramatic decrease in potency with an IC50
of 345 nM (Table 3).
With the accomplishment of these derivatizations, the lead
structure of the C-5 scaffold was identified as the 2-
phenylazaindole 14f.
The nonrigidized counterpart ML3403 (Figure 2) of the 2-
phenylazaindole scaffold, the phenylmethylamine substituted
pyridine, proved to be a highly potent derivative, with the
assumption that the hydrophobic interaction sites were
extended, as provided by the hydrophobic interaction sites
above and below the pyridine ring.9 This beneficial effect was
reduced by our rigidization concept because of the integration
of the methyl residue into the rigidized aromatic system.
The 1YWR crystal was chosen for docking experiments. The
ligand ((S)-4-(4-fluorophenyl)-1-methyl-5-(2-(1-
phenylethylamino)pyrimidin-4-yl)-2-(piperidin-4-yl)-1,2-dihy-
dropyrazol-3-one) bound in X-ray structure 1YWR shows that
the benzyl is nonplanar with the pyrimidine ring (Figure 13).
Figure 13. Overlay of the bound ligand of crystal structure 1YWR25
and a 2-phenylated azaindole analogue of the herein synthesized
inhibitors.
Nevertheless, our scaffold proved to be more efficient than
the nonrigidized analogue (ML3403). In comparison to the
previously determined IC50 values of the nonsubstituted
analogue, only a 2-fold increase in the biological activity was
observed.9 In comparison to our unsubstituted analogue 14d,
we observed an 8-fold increase of the biological activity
(compared by activity coefficients) (Tables 2 and 3).
8435
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
The biological data clearly demonstrate the beneficial effect
of the additional hydrogen bonding (14c) (Table 2), with a 1.4-
fold increase in potency and an exceedingly strong hydrophobic
effect for 14f with a 5-fold increase (Table 3), compared to 14c.
With the 2-phenylazaindole we observed a highly potent
scaffold for interactions with the hinge region and with
hydrophobic region II. With IC50 values as low as 16 nM, we
further optimized our inhibitors by derivatization of the
imidazole core (Figure 14) in stage V. The imidazole C-2
A dramatic increase in potency could be observed by the
structural reduction of the p-methoxy moiety to the more polar
p-hydroxy derivative 20f (Figure 14, Table 4), the structural
analogue of SB202190, with a nearly 20-fold increased potency,
compared to the lead compound SB203580, and a 66-fold
increased potency compared to our incipient reference
compound 14d (Figure 10, Table 2).
The accomplished docking studies of 20f (Figure 15) gave
similar results to that of 14f (Figure 12). The role of the p-
Figure 14. C-2 modifications of the imidazole core (derivatives 15f−
20f, stage V of optimization strategy).
position, bearing a phenyl substituent, has already been shown
to be a less than ideal substituent.3 Therefore, we investigated
several aryl substituents at the imidazole C-2 position,
combined with the most effective C-5 substituent, the
phenylazaindole (Figure 14).
A trimethoxyphenyl substitution 15f (Figure 14, Table 4) led
to a decreased biological activity, while the methoxyphenyl
Table 4. Inhibition of p38 MAPK
a
compd
IC50 [μM]
gradient factor, IC50(SB203580)/IC50(compd)
17f
18f
19f
20f
15f
16f
0.039 0.0081
0.008 0.0003
0.011 0.0003
0.003 0.003
0.290 0.0453
0.070 0.0079
0.97
4.75
3.46
19.67
0.12
0.67
Figure 15. Docking results for 20f in the p38 MAP kinase X-ray
structure 1YWR25 from the PDB database.
hydroxy group could not be elucidated by docking experiments.
A possible explanation could be a participation of bridged water
interactions, which might lead to an enhanced binding to the
receptor.
a
Mean SEM of three experiments. Compounds tested according to
Laufer et al.24
For drug candidates interfering with protein kinases, their
range of specifity within the kinome is a key question, and a
robust evaluation of potency and selectivity is an essential part
of drug discovery. Rigidization is a promising strategy in
modern medicinal chemistry. Its objective is to synthetically fix
a flexible ligand scaffold into a biologically relevant con-
formation with the overall intention to improve potency and
selectivity of the corresponding ligands.11 The herein evaluated
inhibitors match this concept, and the selectivity of the most
potent inhibitor 20f was investigated for its selectivity toward
other kinases. The corresponding analogue of 20f is the well-
known and already mentioned inhibitor SB202190, which is
superior to SB203580 in regard to both potency and selectivity
as shown by Fabian et al.26 SB202190 was investigated in a
panel of 119 kinases, and the compound showed significant
inhibition of 14 kinases within this panel. Relevant inhibition
was observed on kinases CSNK1E, EGFR, FRK, GAK, JNK1,
JNK2, JNK3, p38α, p38β, p38γ, PTK6, RIPK2, STK36, and
TNIK. In the same panel, compound 20f showed an improved
selectivity profile. Significant inhibition was observed toward
the kinases GAK, CSNK1E, EGFR, p38α, p38β, RIPK2, SRC,
STK36, and TNIK.
derivative 16f displayed a better but still lower activity profile
(Figure 14, Table 4). A meta substitution on the aryl derivative
was poorly tolerated; therefore, we focused on the para-
substituted aryl derivatizations in the following investigations.
The p-methylsulfanyl derivative 17f (Figure 14, Table 4) should
serve as a direct analogue of the methoxy derivative 16f (Figure
14, Table 4), with the effect of lower basicity of the heteroatom.
An improved biological activity compared to the oxygen
analogue was observed, but with respect to the reference
compound 14f (Figure 10, Table 3), the potency was clearly
decreased. The common basis of these modifications is the
lipophilic character of the performed derivatizations; therefore,
the corresponding substitution patterns of the phenyl moiety
were rendered more polar, in analogy to SB203580 and
SB202190.2 The methylsulfanyl was oxidized to the corre-
sponding sulfoxide 18f (Figure 14, Table 4), the structural
analogue of SB203580, and a further oxidation yielded the
corresponding sulfone 19f (Figure 14, Table 4). Both inhibitors
were superior to 14f and SB203580 (Figure 14, Table 4),
whereby the sulfoxide (racemic) proved to be the more potent
derivative with a gradient factor of 4.75.
8436
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
Plesset perturbation theory) calculations are omitted and just DFT
(density functional theory) energies were used.
Table 5. Relevant Inhibition of 20f in a Panel of 119 Kinases
gene symbol (locus link) gene symbol kinase activity [%] at 1000 nM
Methods and Materials. All commercially available reagents and
solvents were used without further purification. NMR data were
obtained on a Bruker Spectrospin AC 200 instrument at ambient
temperature. High-resolution spectral mass data were obtained on a
Thermo Finnigan TSQ70 instrument. Melting points were measured
on a Buechi melting point B-545 instrument. IR spectra were obtained
by a Perkin-Elmer Spectrum One (ATR) instrument. LC−mass
spectrometry (MS) spectra were recorded by a Finnigan TSQ
Quantum triple quadrupole mass spectrometer (Finnigan MAT).
GC−MS was done using a Hewlett-Packard HP 6890 series GC
system (Hewlett-Packard HP 5973 mass selective detector). The
purity of the final compounds was determined by HPLC on a Hewlett-
Packard HP 1090 series II liquid chromatograph using a Betasil C8
column (150 mm × 4.6 mm i.d., dp = 5 μm, Thermo Fisher Scientific,
Waltham, MA) at 230 and 254 nm, employing a gradient of 0.01 M
KH2PO4 (pH 2.3) and methanol as the solvent system with a flow rate
of 1.5 mL/min. All final compounds have a purity of >96% (see
Supporting Information for details).
General Procedure of Stille Coupling. A premilled mixture (0.1
mol % Pd) of Pd(OAc)2 and X-Phos (1/3) was stirred in dioxane for
30 min, and the yellow solution turned to a dark purple solution. The
halogenated N-protected heterocycle and the stannylimidazole (2
equiv) were added to the reaction mixture and heated under
microwave irradiation (300 W, 100 °C, 10−15 min) in a sealed tube
under nitrogen atmosphere and instant cooling. After quenching with
KF (40% in water) and extracting with EtOAc, the crude product was
purified by flash chromatography.
CSNK1E
EGFR
CSNK1E
EGFR
GAK
17
2.8
7.7
8.9
2.5
3.2
28
GAK
MAPK14
MAPK11
RIPK2
p38α
p38β
RIPK2
SRC
SRC
STK36
KIAA0551
STK36
TNIK
4.8
17
Compared to SB202190, 20f showed an improved selectivity
profile in a panel of 119 kinases with a significant inhibition of 9
kinases whereas SB202190 showed significant inhibition of 14
kinases. The selectivity of compound 20f was shown to be
markedly improved within two kinase families the MAP kinase
family and the tyrosin kinase family. Within the MAP kinases,
compound 20f showed only significant inhibition of p38α and
p38β, whereas SB202190 shows inhibition of JNK1, JNK2,
JNK3, p38α, p38β, and p38γ. Within the tyrosin kinase family,
20f showed a completely different selectivity profile, inhibiting
the SRC kinase, and SB202190 inhibited the FRK and PTK6
kinases.
A complete list of inhibition data for all 119 kinases is
included in the Supporting Information.
General Procedure for MOM Deprotection. The MOM
protected compound was dissolved in a mixture of aqueous HCl/
MeOH (4/1). The reaction mixture was heated under reflux until
complete conversion to the product (monitoring by HPLC). After
removal of the solvent, the crude product was neutralized with
NaHCO3 and extracted with ethyl acetate and subjected to flash
chromatography.
CONCLUSION
■
Despite the extensive research of the class of pyridinylimida-
zoles as p38 MAPK inhibitors, no known investigation has
focused on the concept of rigidization/aromatization. With the
present inhibitor design, a fundamental structural change was
successfully accomplished, leading to highly active inhibitors
with IC50 values as low as 3 nM and a markedly improved
selectivity profile. This study has given a platform to further
investigate the pharmacological aspects of these promising
modifications, based on a highly flexible and efficient synthetic
procedure to provide further complex C-5 variations of the
imidazole core.
General Procedure for Detosylation. The educt was dissolved
in MeOH. After addition of Cs2CO3, the mixture was heated to reflux
until complete conversion to the product. The solvent was removed,
and water was added to the crude product. After extraction with ethyl
acetate, the crude product was subjected to flash chromatography.
6-[4-(4-Fluorophenyl)-1-(methoxymethyl)-2-phenyl-1H-imi-
dazol-5-yl]-9H-purine (15a). Yield, 46%; C22H17FN6O (Mr
=
400.41); 1H NMR (200 MHz, acetone-d6) δ ppm 2.89 (s, 3 H,
MOMCH3), 5.59 (s, 2 H, MOMCH2), 6.92 (m, 2 H, m-4FPhH), 7.53
(m, 6 H, o-/m-/p-PhH, Pyr.6-H, PurH), 7.90 (m, 2 H, o-4FPhH), 8.38
(s, 1 H, Pur.2-H), 9.03 (m, 1 H, NH); 13C NMR (50.33 MHz,
acetone-d6) δ ppm 555.0, 75.4, 114.5 (d, JC−F = 21.5 Hz), 123.7, 125.6,
128.3, 128.5, 128.6, 128.7, 129.1 (d, JC−F =8.1 Hz), 129.2, 130.5, 131.0
(d, JC−F =3.1 Hz), 131.6, 131.7, 131.8, 131.8, 141.0, 145.3, 150.2, 161.9
(d, JC−F = 244.0 Hz); HRMS (ESI) calcd for C22H17FN6O [M + H]+
401.152 06, found 401.151 701 (error 0.90 ppm).
EXPERIMENTAL SECTION
■
Molecular Modeling. All modeling was performed on an Ubuntu
10.04 LTS system. For visualization and building the structures,
Maestro from Schrodinger (Maestro, version 9.1, Schrodinger, LLC,
New York, NY, 2010) was used. The illustrations of modeling were
generated by PyMol (version 0.99rc6). The docking studies were
̈
̈
4-[4-(4-Fluorophenyl)-1-(methoxymethyl)-2-phenyl-1H-imi-
dazol-5-yl]-7H-pyrrolo[2,3-d]pyrimidine (15b). Yield, 42%;
1
performed using the Induced Fit Docking protocol from Schrodinger
(Schrodinger Suite 2010 Induced Fit Docking protocol; Glide, version
̈
C23H18FN5O (Mr = 399,42); H NMR (200 MHz, methanol-d4) δ
̈
ppm 2.79 (s, 3 H, MOMCH3), 5.46 (s, 2 H, MOMCH2), 5.89 (d, J =
3.41 Hz, 1 H, Pyr.6-H), 6.87 (m, 2 H, m-4FPhH), 7.29 (m, 3 H, o-
4FPhH, Pyr.5H), 7.49 (m, 3 H, m-/p-PhH), 7.75 (m, 2 H, o-PhH),
8.83 (s, 1 H, NH); 13C NMR (50.33 MHz, methanol-d4) δ ppm 54.7,
75.4, 100.1, 114.7 (d, JC−F = 21.9 Hz), 117.7, 125.1, 127.4, 128.4,
5.6, Schrodinger, LLC, New York, NY, 2010; Prime, version 2.2,
̈
Schrodinger, LLC, New York, NY, 2010). The PDB codes and
̈
references for the used X-ray structures are mentioned in the caption
of the figures. In all docking studies mentioned here, the SP precision
was used. The presented docking results were ranked highest.
The best docking pose of the corresponding ligand was minimized
129.3, 129.4, 129.7, 140.6, 148.6, 150.5, 151.2, 152.0, 162.3 (d, JC−F
=
246.2 Hz); HRMS (ESI) calcd for C23H18FN5O [M + H]+ 400.156 81,
found 400.156 528 (error 0.71 ppm).
with MacroModel (MacroModel, version 9.8, Schrodinger, LLC, New
̈
York, NY, 2010) using the default settings (force field, OPLS 2005;
solvent, water) and frozen atoms as constraints. After the minimization
the hydrogen bond energy calculation was done using the Jaguar batch
4-[4-(4-Fluorophenyl)-1-(methoxymethyl)-2-phenyl-1H-imi-
dazol-5-yl]-1H-pyrrolo[2,3-b]pyridine (15c). Yield, 96%;
1
C24H19FN4O (Mr = 398.43); H NMR (200 MHz, CDCl3) δ ppm
script hydrogen_bond.py (Jaguar, version 7.7, Schrodinger, LLC, New
̈
3.05 (m, 3 H, MOMCH3), 5.10 (m, 2 H, MOMCH2), 6.11 (m, 1 H,
Aza2-H), 6.91 (m, 2 H, m-4FPhH), 7.36 (m, 5 H, m-/p-PhH, Aza3-
H), 7.58 (m, 3 H, Aza5-H, o-4FPhH), 7.89 (m, 2 H, o-PhH), 8.35 (m,
1 H, Aza6-H); 13C NMR (50.33 MHz, methanol-d4) δ ppm 54.5, 75.0,
York, NY, 2010) which is integrated in the Schrodinger software suite.
̈
During this calculation there was no optimization for any structure and
the fast mode was chosen. In the fast mode LMP2 (local Møller−
8437
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
99.6, 114.5 (d, JC−F = 21.8 Hz), 117.1, 120.3, 120.3, 1265, 128.4, 126.7
(d, JC−F = 8.1 Hz), 129.0, 129.4, 129.5, 128.8 (d, JC−F = 3.1 Hz), 130.3,
137.9, 142.0, 148.3, 162.0 (d, JC−F = 245.8 Hz); HRMS (ESI) calcd for
C24H19FN4O [M + H]+ 399.161 57, found 399.161 961 (error 0.98
ppm).
1.84 (m, 4 H, ChH), 2.69 (m, 1 H, ipso-ChH), 6.04 (m, 1 H, Aza3-H),
6.95 (m, 2 H, m-4FPhH), 7.13 (d, J = 5.43 Hz, 1 H, Aza5-H), 7.46 (m,
6 H, m-/p-PhH, o-4FPhH, NH), 7.95 (m, 1 H, Aza6-H), 8.04 (m, 2 H,
o-PhH), 10.68 (s, 1 H, NH); 13C NMR (50.33 MHz, DMSO-d6) δ
ppm 25.6, 31.9, 33.9, 36.5, 45.2, 95.6, 114.7, 115.1 [(d, JC−F = 21.47
Hz), 118.0, 125.5, 128.6 (d, JC−F = 8.06 Hz), 128.7, 129.7, 129.8 (br),
129.9, 140.7, 146.1, 149.0, 161.4 (d, JC−F = 245.0 Hz); HRMS (ESI)
calcd for C28H25FN4 [M + H]+ 437.213 60, found 437.213 892 (error
0.67 ppm).
4-[4-(4-Fluorophenyl)-2-(3,4,5-trimethoxyphenyl)-1H-imida-
zol-5-yl]-2-phenyl-1H-pyrrolo[2,3-b]pyridine (15f). Yield, 84%;
C31H25FN4O3 (Mr = 520.55); 1H NMR (200 MHz, DMSO-d6) δ ppm
3.71 (s, 3 H, p-OCH3), 3.87 (s, 6 H, m-OCH3), 6.60 (m, 1 H, Aza3-
H), 7.12 (m, 2 H, m-4FPhH), 7.41 (m, 8 H, Aza5-H, o-ArH, o-4FPhH,
m-/p-4FPhH), 7.76 (m, 2 H, o-PhH), 8.23 (m, 1 H, Aza6-H), 12.18
(m, 1 H, NH), 12.88 (m, 1 H, NH); 13C NMR (50.33 MHz, DMSO-
d6) δ ppm 56.0, 60.1, 97.8, 102.8, 115.3 (d, JC−F = 20.7 Hz), 118.7,
125.2, 125.6, 128.0, 128.9, 129.7 (d, JC−F = 3.1 Hz), 131.5, 137.9,
142.8, 146.2, 150.4, 153.2, 161.4 (d, JC−F = 245.8 Hz); HRMS (ESI)
calcd for C31H25FN4O3 [M + H]+ 521.198 35, found 521.198 712
(error 0.70 ppm).
4-[4-(4-Fluorophenyl)-2-(4-methoxyphenyl)-1H-imidazol-5-
yl]-2-phenyl-1H-pyrrolo[2,3-b]pyridine (16f). Yield, 75%;
C29H21FN4O (Mr = 460.5); 1H NMR (200 MHz, DMSO-d6/
methanol-d4) δ ppm 3.90 (s, 3 H, OCH3), 6.49 (brm, 1 H, Aza-3-
H), 7.10 (m, 4 H, m-4FPhH, m-ArH), 7.23 (d, J = 5.05 Hz, 1 H, Aza5-
H), 7.45 (m, 5 H, m-/p-PhH, o-4FPhH), 7.72 (m, 2 H, o-PhH), 8.05
(d, J = 8.97 Hz, 2 H, o-ArH), 8.23 (brs, 1 H, Aza6-H); HRMS (ESI)
calcd for C29H21FN4O [M + H]+ 461.177 22, found 461.177 353
(error 0.29 ppm).
4-{4-(4-Fluorophenyl)-2-[4-(methylthio)phenyl]-1H-imida-
zol-5-yl}-2-phenyl-1H-pyrrolo[2,3-b]pyridine (17f). Yield, 69%;
C29H21FN4S (Mr = 476.57); 1H NMR (200 MHz, methanol-d4/
DMSO-d6) δ ppm 2.45 (s, 3 H, SCH3), 6.46 (s, 1 H, Aza3-H), 7.05
(m, 3 H, m-4FPhH, Aza5-H), 7.29 (m, 5 H, m-PhSCH3, m-/p-PhH),
7.45 (m, 2 H, o-4FPhH), 7.64 (d, J = 7.07 Hz, 2 H, m-PhH), 7.97 (d, J
= 8.34 Hz, 2 H, o-PhSCH3), 8.12 (d, J = 5.05 Hz, 1 H, Aza6-H); 13C
NMR (50.33 MHz, methanol-d4/DMSO-d6) δ ppm 99.3, 116.5 (d,
JC−F =21.4 Hz), 116.9, 120.6, 121.1, 126.7, 127.4 (d, JC−F = 8.4 Hz),
127.9, 129.4, 130.2, 131.4 (d, JC−F = 4.9 Hz), 133.1, 139.9, 141.2,
143.9, 147.0, 148.2, 151.7, 163.1 (d, JC−F = 245.8 Hz); HRMS (ESI)
calcd for C29H21FN4S [M + H]+ 477.154 37, found 477.154 424 (error
0.11 ppm).
6-[4-(4-Fluorophenyl)-2-phenyl-1H-imidazol-5-yl]-9H-purine
1
(14a). Yield, 57%; C20H13FN6 (Mr = 356.36); H NMR (200 MHz,
methanol-d4/DMSO-d6) δ ppm 7.14 (m, 2 H, m-4FPhH), 7.42 (m, 3
H, m-/p-4FPhH), 7.82 (m, 2 H, o-4FPhH), 8.17 (d, J = 6.82 Hz, 2 H,
o-PhH), 8.47 (s, 1 H, Pur.8-H), 8.66 (s, 1 H, Pur.2-H); 13C NMR
(100.66 MHz, DMSO-d6) δ ppm 114.6 (d, JC−F = 22.0 Hz), 125.9,
128.7, 129.0, 129.5, 132.0, 145.6, 146.8, 147.3, 151.5, 162.0 (d, JC−F
=
245.9 Hz); HRMS (ESI) calcd for C20H13FN6 [M + H]+ 357.1285,
found 357.125 623 (error 0.64 ppm).
4-[4-(4-Fluorophenyl)-2-phenyl-1H-imidazol-5-yl]-7H-
pyrrolo[2,3-d]pyrimidine (14b). Yield, 85%; C21H14FN5 (Mr
=
1
355.37); H NMR (200 MHz, DMSO-d6) δ ppm 5.86 (s, 1 H, Pyr.5-
H), 7.32 (m, 2 H, m-4FPhH), 7.63 (m, 7 H, m-/p-PhH, o-4FPhH,
Pyr.6-H), 8.27 (m, 2 H, o-PhH), 8.99 (m, 1 H, Pyr.2-H), 13.17 (m, 1
H, NH); 13C NMR (50.33 MHz, DMSO-d6) δ ppm 103.1, 113.6,
116.1 (d, JC−F = 21.5 Hz), 125.6, 126.7, 127.8, 128.4 (d, JC−F = 4.2
Hz), 129.3, 129.3, 130.5 (d, JC−F = 8.8 Hz), 132.0, 132.1, 136.4, 136.5,
145.6, 146.5, 148.0, 152.5, 168.2 (d, JC−F = 273.8 Hz); HRMS (ESI)
calcd for C21H14FN5 [M + H]+ 356.130 60, found 356.130 892 (error
0.82 ppm).
4-[4-(4-Fluorophenyl)-2-phenyl-1H-imidazol-5-yl]-1H-
pyrrolo[2,3-b]pyridine (14c). Yield, 82%; C22H15FN4 (Mr
=
1
354.38); H NMR (200 MHz, methanol-d4) δ ppm 6.08 (m, 1 H,
Aza2-H), 6.94 (m, 2 H, o-4FPhH), 7.11 (d, J = 5.05 Hz, 1 H, Aza5-H),
7.18 (d, J = 3.54 Hz, 1 H, Aza3-H), 7.38 (m, 6 H, m-/p-PhH, o-
4FPhH), 7.94 (m, 2 H, o-PhH), 8.10 (d, J = 5.31 Hz, 1 H, Aza6-H); 13
C NMR (50.33 MHz, methanol-d4) δ ppm 100.5, 114.8 (d, JC−F = 21.5
Hz), 118.5, 125.4, 125.4, 125.5, 128.5, 128.7, 129.6 (d, JC−F = 6.1 Hz),
129.7, 141.6, 147.6, 148.3, 162.2 (d, JC−F = 246.0 Hz); HRMS (ESI)
calcd for C22H15FN4 [M + H]+ 355.135 35, found 355.135 511 (error
0.45 ppm).
4-[4-(4-Fluorophenyl)-2-phenyl-1H-imidazol-5-yl]-9H-
pyrido[2,3-b]indole (14e). Yield, 69%; C26H17FN4 (Mr = 404.44);
1H NMR (200 MHz, methanol-d4) δ ppm 6.86 (m, 3 H, m-4FPhH),
7.08 (d, J = 4.55 Hz, 1 H, Carb.H), 7.40 (m, 8 H, m-/p-PhH, o-
4FPhH), 7.95 (m, 2 H, o-PhH), 8.27 (m, 1 H, Carb.2-H); 13C NMR
(50.33 MHz, methanol-d4) δ ppm 112.3, 116.5 (d, JC−F = 22.2 Hz),
117.9, 121.1, 121.3, 123.7, 127.3, 128.4, 129.7, 130.3, 130.5 (d, JC−F
=
8.4 Hz), 131.1, 134.4, 136.2, 140.8, 145.9, 151.0 (d, JC−F = 230.4 Hz);
HRMS (ESI) calcd for C26H17FN4 [M + H]+ 405.151 00, found
405.150 678 (error 0.80 ppm).
4-[4-(4-Fluorophenyl)-5-(2-phenyl-1H-pyrrolo[2,3-b]pyridin-
4-yl)-1H-imidazol-2-yl]phenol (20f). Yield, 88%; C28H19FN4O (Mr
1
= 446.48); H NMR (200 MHz, methanol-d4/DMSO-d6) δ ppm 6.74
(s, 1 H, Aza3-H), 7.15 (d, J = 8.59 Hz, 2 H, m-PhOH), 7.36 (m, 2 H,
m-4FPhH), 7.46 (d, J = 5.05 Hz, 1 H, Aza5-H), 7.68 (m, 5 H, o-
4FPhH, m-PhH, p-PhH), 7.96 (d, J = 7.07 Hz, 2 H, m-PhH), 8.17 (d, J
= 8.84 Hz, 2 H, o-PhOH), 8.46 (d, J = 5.31 Hz, 1 H, Aza6-H); 13C
NMR (50.33 MHz, methanol-d4/DMSO-d6) δ ppm 99.3, 116.4 (d,
JC−F = 21.9 Hz), 116.9, 117.2, 120.6, 122.0, 126.6, 128.7, 129.3, 130.2,
131.3 (d, JC−F = 8.1 Hz), 133.1, 139.8, 143.8, 149.5, 151.6, 163.4 (d,
JC−F = 245.8 Hz); HRMS (ESI) calcd for C28H19FN4O [M + H]+
447.161 57, found 447.161 709 (error 0.31 ppm).
2-Benzyl-4-[4-(4-fluorophenyl)-2-phenyl-1H-imidazol-5-yl]-
1H-pyrrolo[2,3-b]pyridine (14g). Yield, 65%; C29H21FN4 (Mr =
444.5); 1H NMR (400 MHz, acetone-d6/methanol-d4) δ ppm 3.93 (s,
2 H, BzCH2), 6.96 (m, 2 H, Aza3-H/Aza5-H), 7.06 (m, 2 H, o-BzH),
7.18 (m, 5 H, m-/p-BzH, m-4FPhH), 7.35 (m, 3 H, m-/p-ArH), 7.44
(m, 2 H, o-4FPhH), 7.98 (d, J = 7.33 Hz, 2 H, o-PhH), 8.09 (m, 1
H,Aza6-H). 13C could not be detected because of insolubility of the
compound. HRMS (ESI) calcd for C29H21FN4 [M + H]+ 445.182 30,
found 445.182 466 (error 0.37 ppm).
4-[4-(4-Fluorophenyl)-2-phenyl-1H-imidazol-5-yl]-2-phenyl-
1H-pyrrolo[2,3-b]pyridine (14f). Yield, 90%; C28H19FN4 (Mr =
430.48); 1H NMR (200 MHz, CDCl3) δ ppm 6.80 (s, 1 H, NH), 7.10
(m, 2 H, o-4FPhH), 7.21 (d, J = 4.80 Hz, 1 H, Aza3-H), 7.42 (brm, 7
H, PhH, Aza5-H), 7.63 (m, 2 H, m-4FPhH), 7.86 (d, J = 7.20 Hz, 2 H,
o-PhH), 8.14 (d, J = 8.59 Hz, 1 H, Aza6-H), 8.22 (m, 2 H, o-PhH),
11.85 (m, 1 H, NH); 13C NMR (50.33 MHz, acetone-d6) 98.5, 114.97
(d, JC−F = 21.3 Hz), 119.3, 125.4, 125.8, 126.0, 127.1, 128.0, 128.5 (d,
JC−F = 8.5 Hz), 128.9, 129.8, 130.5, 132.0, 138.5, 139.5, 142.6, 146.4,
146.7, 150.8, 162 (d, JC−F = 244.1 Hz); HRMS (ESI) calcd for
C28H19FN4 [M + H]+ 431.166 65, found 431.166 381 (error 0.62
ppm).
4-{4-(4-Fluorophenyl)-2-[4-(methylsulfinyl)phenyl]-1H-imi-
dazol-5-yl}-2-phenyl-1H-pyrrolo[2,3-b]pyridine (18f). Yield,
1
57%; C29H21FN4OS (Mr = 492.57); H NMR (200 MHz, methanol-
d4) δ ppm 2.84 (s, 3 H, SOCH3), 6.45 (s, 1 H, Aza3H), 7.07 (m, 2 H,
m-4FPhH), 7.21 (d, J = 5.05 Hz, 1 H, Aza 5-H), 7.40 (m, 5 H, m-/p-
PhH, o-4FPhH), 7.65 (m, 2 H, o-PhH), 7.83 (d, J = 8.72 Hz, 2 H, m-
ArH), 8.17 (m, 1 H, Aza6-H), 8.26 (d, J = 8.72 Hz, 2 H, o-ArH); 13C
NMR (50.33 MHz, methanol-d4/DMSO-d6) δ ppm 43.5, 99.1,110.8,
116.5 (d, JC−F = 21.9 Hz), 119.6, 125.6, 126.6, 127.9, 129.3, 123.0,
131.4 (d, JC−F = 8.4 Hz), 133.1, 134.1, 140.6, 142.4, 143.3, 146.4,
147.6, 151.2, 164.0 (d, JC−F = 246.9 Hz); HRMS (ESI) calcd for
C29H21FN4OS [M + H]+ 493.149 29, found 493.149 568 (error 0.58
ppm).
2-Cyclohexyl-4-[4-(4-fluorophenyl)-2-phenyl-1H-imidazol-5-
yl]-1H-pyrrolo[2,3-b]pyridine (14h). Yield, 85%; C28H25FN4 (Mr =
4-{4-(4-Fluorophenyl)-2-[4-(methylsulfonyl)phenyl]-1H-imi-
dazol-5-yl}-2-phenyl-1H-pyrrolo[2,3-b]pyridine (19f). Yield,
1
436.52); H NMR (200 MHz, CDCl3) δ ppm 1.29 (m, 6 H, ChH),
8438
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439

Journal of Medicinal Chemistry
Article
1
50%; C29H21FN4O2S (Mr = 508.57); H NMR (400 MHz, methanol-
(10) Laufer, S. A.; Wagner, G. K.; Kotschenreuther, D. A.; Albrecht,
W. Novel substituted pyridinyl imidazoles as potent anticytokine
agents with low activity against hepatic cytochrome P450 enzymes. J.
Med. Chem. 2003, 46, 3230−3244.
d4) δ ppm 3.04 (s, 3 H, SO2CH3), 6.54 (s, 1 H, aza3-H), 6.96 (m, 2 H,
m-4FPhH), 7.12 (m, 1 H), 7.23 (d, J = 7.33 Hz, 1 H, Aza5-H), 7.31
(m, 2 H, o-4FPhH), 7.41 (m, 2 H, o-PhH), 7.56 (m, 3 H, m-/p-PhH),
7.94 (d, J = 8.34 Hz, 1 H, o-PhSO2H), 8.05 (m, 1 H, aza6-H), 8.22 (d,
J = 8.34 Hz, 2 H, m-PhSO2H); 13C NMR (200 MHz, DMSO-d6) δ
ppm 43.2, 96.2, 115.3 (d, JC−F = 21.5 Hz), 117.9, 120.6, 125.6, 127.2,
128.2, 128.3, 128.5, 128.9, 129.4 (d, JC−F = 3.1 Hz), 129.8, 131.0,
136.8, 140.0, 143.1, 150.1, 161.4 (d, JC−F = 244.6 Hz); HRMS (ESI)
calcd for C29H21FN4O2S [M + H]+ 509.144 20, found 509.144 424
(error 0.44 ppm).
(11) The Practice of Medicinal Chemistry; Wermuth, C. G., Ed.;
Elsevier: Amsterdam, The Netherlands, 2008.
(12) Velazquez-Campoy, A.; Luque, I.; Freire, E. The application of
thermodynamic methods in drug design. Thermochim. Acta 2001, 380,
217−227.
(13) Veber, D. F.; Johnson, S. R.; Cheng, H. Y.; Smith, B. R.; Ward,
K. W.; Kopple, K. D. Molecular properties that influence the oral
bioavailability of drug candidates. J. Med. Chem. 2002, 45, 2615−2623.
(14) Selig, R.; Schollmeyer, D.; Albrecht, W.; Laufer, S. The
application of Stille cross-coupling reactions with multiple nitrogen
containing heterocycles. Tetrahedron 2011, 67, 9204−9213.
(15) Hartwig, J. F. Discovery and understanding of transition-metal-
catalyzed aromatic substitution reactions. Synlett 2006, 1283−1294.
(16) Robins, R. K.; Godefroi, E. F.; Taylor, E. C.; Lewis, L. R.;
Jackson, A. Purine nucleosides. I. The synthesis of certain 6-
substituted-9-(tetrahydro-2-pyranyl)purines as models of purine
deoxynucleosides. J. Am. Chem. Soc. 1961, 83, 2574−2579.
(17) Mendiola, J.; Baeza, A.; varez-Builla, J.; Vaquero, J. J. Reaction of
bromomethylazoles and tosylmethyl isocyanide. A novel heterocycliza-
tion method for the synthesis of the core of marine alkaloids variolins
and related azolopyrimidines. J. Org. Chem. 2004, 69, 4974−4983.
(18) Liverton, N. J.; Butcher, J. W.; Claiborne, C. F.; Claremon, D.
A.; Libby, B. E.; Nguyen, K. T.; Pitzenberger, S. M.; Selnick, H. G.;
Smith, G. R.; Tebben, A.; Vacca, J. P.; Varga, S. L.; Agarwal, L.;
Dancheck, K.; Forsyth, A. J.; Fletcher, D. S.; Frantz, B.; Hanlon, W. A.;
Harper, C. F.; Hofsess, S. J.; Kostura, M.; Lin, J.; Luell, S.; O’Neill, E.
A.; Orevillo, C. J.; Pang, M.; Parsons, J.; Rolando, A.; Sahly, Y.; Visco,
D. M.; O’Keefe, S. J. Design and synthesis of potent, selective, and
orally bioavailable tetrasubstituted imidazole inhibitors of p38
mitogen-activated protein kinase. J. Med. Chem. 1999, 42, 2180−2190.
(19) Ahlquist, M.; Fristrup, P.; Tanner, D.; Norrby, P. O. Theoretical
evidence for low-ligated palladium(0): [Pd-L] as the active species in
oxidative addition reactions. Organometallics 2006, 25, 2066−2073.
(20) Cox, P. J.; Majid, T. N.; Lai, J. Y. Q.; Morley, A. D.; Amendola,
S.; Deprets, S. Preparation of Azaindoles as Protein Kinase Inhibitors.
U.S. Patent WO2001047922, 2000.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures, analytical data, detailed selectivity
data, and the molecular modeling. This material is available free
of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +49-70712978788. Fax: +49-7071295037. E-mail:
stefan.laufer@uni-tuebingen.de.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the Federal Ministry of Education and
Research, Germany, Merckle GmbH, Ulm, Germany, the BASF
Group, especially Dr. Martin Fiene, and the Fonds der
Chemischen Industrie, Germany, for their generous support
of this work.
REFERENCES
■
(1) Lee, J. C.; Griswold, D. E.; Votta, B.; Hanna, N. Inhibition of
monocyte IL-1 production by the anti-inflammatory compound, SK &
F 86002. Int. J. Immunopharmacol. 1988, 10, 835−843.
(21) Maresh, J. J.; Giddings, L. A.; Friedrich, A.; Loris, E. A.; Panjikar,
S.; Trout, B. L.; Stoeckigt, J.; Peters, B.; O’Connor, S. E. Strictosidine
synthase: mechanism of a Pictet−Spengler catalyzing enzyme. J. Am.
Chem. Soc. 2008, 130, 710−723.
(22) Bajwa, J. S.; Chen, G. P.; Prasad, K.; Repic, O.; Blacklock, T. J.
Deprotection of N-tosylated indoles and related structures using
cesium carbonate. Tetrahedron Lett. 2006, 47, 6425−6427.
(23) Sasaki, Y.; Kato, D.; Boger, D. L. Asymmetric total synthesis of
vindorosine, vindoline, and key vinblastine analogues. J. Am. Chem. Soc.
2010, 132, 13533−13544.
(24) Laufer, S.; Thuma, S.; Peifer, C.; Greim, C.; Herweh, Y.;
Albrecht, W.; Dehner, F. An immunosorbent, nonradioactive p38
MAP kinase assay comparable to standard radioactive liquid-phase
assays. Anal. Biochem. 2005, 344, 135−137.
(25) Golebiowski, A.; Townes, J. A.; Laufersweiler, M. J.; Brugel, T.
A.; Clark, M. P.; Clark, C. M.; Djung, J. F.; Laughlin, S. K.; Sabat, M.
P.; Bookland, R. G.; VanRens, J. C.; De, B.; Hsieh, L. C.; Janusz, M. J.;
Walter, R. L.; Webster, M. E.; Mekel, M. J. The development of
monocyclic pyrazolone based cytokine synthesis inhibitors. Bioorg.
Med. Chem. Lett. 2005, 15, 2285−2289.
(2) Adams, J. L.; Gallagher, T. F.; Lee, J. Imidazole Derivatives and
Their Use as Cytokine Inhibitors. U.S. Patent WO93114081, 1993.
(3) Peifer, C.; Kinkel, K.; Abadleh, M.; Schollmeyer, D.; Laufer, S.
From five- to six-membered rings: 3,4-diarylquinolinone as lead for
novel p38MAP kinase inhibitors. J. Med. Chem. 2007, 50, 1213−1221.
(4) Gallagher, T. F.; Seibel, G. L.; Kassis, S.; Laydon, J. T.;
Blumenthal, M. J.; Lee, J. C.; Lee, D.; Boehm, J. C.; Fier-Thompson, S.
M.; Abt, J. W.; Soreson, M. E.; Smietana, J. M.; Hall, R. F.; Garigipati,
R. S.; Bender, P. E.; Erhard, K. F.; Krog, A. J.; Hofmann, G. A.;
Sheldrake, P. L.; McDonnell, P. C.; Kumar, S.; Young, P. R.; Adams, J.
L. Regulation of stress-induced cytokine production by pyridinylimi-
dazoles; inhibition of CSBP kinase. Bioorg. Med. Chem. 1997, 5, 49−
64.
(5) 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.
Structural basis of inhibitor selectivity in MAP kinases. Structure 1998,
6, 1117−11128.
(6) Selig, R.; Schattel, V.; Goettert, M.; Schollmeyer, D.; Albrecht,
W.; Laufer, S. Conformational effects on potency of thioimidazoles
and dihydrothiazolines. MedChemComm 2011, 2, 261−269.
(7) Margutti, S.; Laufer, S. A. Are MAP kinases drug targets? Yes, but
difficult ones. ChemMedChem 2007, 2, 1116−1140.
(26) Fabian, M. A.; Biggs, W. H.; Treiber, D. K.; Atteridge, C. E.;
Azimioara, M. D.; Benedetti, M. G.; Carter, T. A.; Ciceri, P.; Edeen, P.
T.; Floyd, M.; Ford, J. M.; Galvin, M.; Gerlach, J. L.; Grotzfeld, R. M.;
Herrgard, S.; Insko, D. E.; Insko, M. A.; Lai, A. G.; Lelias, J. M.; Mehta,
S. A.; Milanov, Z. V.; Velasco, A. M.; Wodicka, L. M.; Patel, H. K.;
Zarrinkar, P. P.; Lockhart, D. J. A small molecule-kinase interaction
map for clinical kinase inhibitors. Nat. Biotechnol. 2005, 23, 329−336.
(8) 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. 1-Substituted 4-aryl-5-pyridinylimidazoles: a new class of cytokine
suppressive drugs with low 5-lipoxygenase and cyclooxygenase
inhibitory potency. J. Med. Chem. 1996, 39, 3929−3937.
(9) Wagner, G.; Laufer, S. Small molecular anti-cytokine agents. Med.
Res. Rev. 2006, 26, 1−62.
8439
dx.doi.org/10.1021/jm300852w | J. Med. Chem. 2012, 55, 8429−8439