p38α mitogen-activated protein kinase inhibitors: Optimization of a series of biphenylamides to give a molecule suitable for clinical progression

Article abstract of DOI:10.1021/jm9004779

p38α MAP kinase is a key anti-inflammatory target for rheumatoid arthritis, influencing biosynthesis of pro-inflammatory cytokines TNFα and IL-1β at a translational and transcriptional level. In this paper, we describe how we have optimized a series of no

Full text of DOI:10.1021/jm9004779

J. Med. Chem. 2009, 52, 6257–6269 6257
DOI: 10.1021/jm9004779
p38r Mitogen-Activated Protein Kinase Inhibitors: Optimization of a Series of Biphenylamides
to Give a Molecule Suitable for Clinical Progression
Nicola M. Aston, Paul Bamborough, Jacqueline B. Buckton, Christopher D. Edwards, Duncan S. Holmes,
Katherine L. Jones, Vipulkumar K. Patel, Penny A. Smee, Donald O. Somers, Giovanni Vitulli, and Ann L. Walker*
GlaxoSmithKline, Medicines Research Centre, Gunnels Wood Rd, Stevenage, Hertfordshire SG1 2NY, U.K.
Received April 14, 2009
p38R MAP kinase is a key anti-inflammatory target for rheumatoid arthritis, influencing biosynthesis of
pro-inflammatory cytokines TNFR and IL-1β at a translational and transcriptional level. In this paper,
we describe how we have optimized a series of novel p38R/β inhibitors using crystal structures of our
inhibitors bound to p38R, classical medicinal chemistry, and modeling of virtual libraries to derive a
molecule suitable for progression into clinical development.
Introduction
Introduction of anticytokine biological therapies, for ex-
ample, Etanercept, Infliximab, Adalimumab, and Anakinra,
which specifically inhibit pro-inflammatory cytokines TNFR
(tumor necrosis factor Ra) and IL-1β (interleukin-1β), have
resulted in significant progress in the treatment of rheumatoid
arthritis (RA) patients who do not respond to current non-
steroidal anti-inflammatories (NSAIDs) or current disease
modifying antirheumatic drugs (DMARDs).1 However, the
administration of these biological therapies by injection or
infusion is burdensome to patients and physicians. An oral
Figure 1. Lead molecule 1 and clinical candidate 2.
therapy, such as a new, small molecule therapy would be a
more convenient administration form. The serine/threonine
1, by increasing potency in an assay to determine the inhibi-
kinase p38R has been identified as a key anti-inflammatory
target, involved in the production of both TNFR and IL-1β,2,3
tion of TNFR production in human whole blood (HWB) and
improving the CYP450 profile resulting in a clinical candidate
2, developed for treatment of rheumatoid arthritis (Figure 1).
and many pharmaceutical companies have pursued p38R
inhibitors as an anti-inflammatory therapy.4-7 To date, none
This molecule is currently also being clinically evaluated for
other inflammatory indications including chronic obstructive
pulmonary disease (COPD) and heart disease.
of these molecules have progressed beyond clinical trials, as
raised levels of liver enzymes associated with liver toxicity
have been seen in some human trials and neurological effects
in others.8,9 Most of the earlier p38R inhibitors have relatively
unselective kinase profiles or show inhibition of cytochrome
p450 isoforms, either of which might be associated with the
side effects seen in the clinic. In our earlier publications,10-13
we described the optimization of a micromolar cross-screen-
ing hit to a small potent and selective p38R/β inhibitor 1 with
an acceptable in vivo pharmacokinetic profile and activity in
the rat peptidoglycan-polysaccharide reactivation (PG-PS)14
and mouse collagen induced arthritis (CIA)15-18 inflamma-
tion models. In this paper, we describe further optimization of
Results and Discussion
Compound 1, while selective for p38R/β and active in rat
reactivation PG-PS (ED50 0.02 mg/kg) and murine CIA (in
which ittotally prevented the progression of arthritis at15mg/
kg/day) models, is weakly active in the HWB assay of inhibi-
tion of TNFR release and displays some potential liabilities
against p450 isoform 2C9 (Table 1). Several potential causes
of the low potency seen in the HWB assay have been con-
sidered; cellular penetration can only be a partial cause as the
drop-off in activity from the enzyme to HWB is larger than
that for the enzyme to the isolated human monocyte assay
(PBMC). To try to understand this difference, a set of 48
compounds, from the biphenyl amide (BPA) and other che-
mical series, was profiled in PBMC and HWB assays using
cells from the same donors. In addition, the physicochemical
properties, CHI LogD (pH2, 7.4, 10.5),19 human serum
albumin (HSA) binding,20 and solubility21 were measured
and further parameters calculated. Statistical models were
*To whom correspondence should be addressed. Phone: 441 438 763
606. Fax: 441 438 763 616. E-mail: Ann.L.Walker@GSK.com.
a Abbreviations: MAP, kinase mitogen-activated protein kinase;
TNFR, tumour necrosis factor alpha; IL-1β, Interleukin 1 beta; RA,
rheumatoid arthritis; DMARD, disease modifying antirheumatic drug;
NSAID, nonsteroidal anti-inflammatory; PG-PS, peptidoglycan-poly-
saccharide; CIA, collagen induced arthritis; COPD, chronic obstructive
pulmonary disease; PBMC, peripheral blood mononuclear cells; HWB,
human whole blood; HSA, human serum albumin; LCMS, liquid
chromatography-mass spectrometry.
r
2009 American Chemical Society
Published on Web 09/22/2009
pubs.acs.org/jmc

6258 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20
Table 1. Effect of Additional Substituents at the 5- and 4-Positions
Aston et al.
compd
R1
R2
p38R FP pKi (n)
PBMC pIC50 (n)
HWB pIC50 (n)
CYP450, 2C9, pIC50
1
3
4
5
6
H
H
H
H
H
F
7.9 (11)
8.4 (6)
7.9 (2)
7.0 (1)
6.9 (2)
6.7 (24)
7.0 (13)
6.8 (2)
nda
5.8 (22)
6.6 (12)
5.9 (2)
nda
5.4
5.7
nda
nda
nda
F
Cl
Me
H
nda
nda
a nd, not determined.
this, a 2500-member virtual library was created, varying the
40-amide. The amines had a wide variety of structural sub-
types, as we knew from previous arrays that p38R has a wide
tolerance in this area of the binding site.10,11 The amines were
chosen on molecular weight and by avoidance of reactive
groups, with no bias from the current preferred groups. To
this library, series specific predictors for oral bioavailability
(high oral absorption - ACDlogD < 4 and CMR <12) and
cytochrome p450 liability [negligible CYP2C9 inhibition
- ACDlogD<4 and ApKa>10 (first acidic pKa)], which
had been generated from analysis of the SAR of previous
examples of the series, were applied.22 As the result of this, a
set of cyclopropylmethyl replacements was chosen and further
87 analogues of 3 prepared. This resulted in a small set of
molecules 7-10 with improved cytochrome p450 profiles and
HWB pIC50>6.5 (Table 2).
Figure 2. Compound 3 bound to p38R showing the fluorine bind-
ing in a small pocket formed by Leu 104 and Lys 53.
then generated on thisdata and testedbyattemptingtopredict
the drop-off for a further set of compounds. The models
proved to be of limited predictive use, partially due to high
interdonor variability for the cellular assays, and no unequi-
vocal understanding was gained regarding factors affecting
the relative PBMC and HWB potencies.
Table 2. Results of the 40-Amide Array Derived by a Virtual Array
Process
Comparison of crystal structures of 1 and a second series of
in-house p38R inhibitors bound to p38R led to the introduc-
tion of a fluorine at the 5-position 3, resulting in a 3-fold
increase in p38R enzyme activity and a similar increase in
PBMC potency but a 6-fold increase in HWB potency.
Replacement of the fluorine by chlorine 4 gave no enhance-
ment in potency in either the enzyme or cellular assays, and
replacement by methyl 5 resulted in a 25-fold drop in enzyme
potency. Shifting the fluorine to the 4-position 6, which
modeling had suggested would be equally valid, resulted in a
9-fold decrease in enzyme potency (Table 1).
These results can be explained by examination of a model
of compound 3 bound to p38R, where it can be seen that the
5-substituent would bind in a lipophilic pocket formed by Leu
104 and Lys 53 (Figure 2). Fluorine appears to be optimal for
the available space, with the largerchlorineand methyl groups
causing unfavorable interactions due to the proximity to the
protein. The lower affinity of 6 can also be explained by
postulating that the 4-fluorine would be electrostatically
unfavorable due to the close proximity of Glu71, approxi-
˚
mately 2.2 A by modeling.
Although 3 has improved HWB potency, it still has a
potential cytochrome p450 liability at 2C9. To overcome

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20 6259
Table 3. Results of 4-Amide Arrays in the Pyridine Series
Figure 3. Compound 11 bound to p38R showing interaction of
pyridine with network of conserved water molecules.
Figure 4. Structures 11-13.
In work ongoing in parallel, comparison of several p38R cry-
stal structures generated for this series suggested that a network
of water molecules fills part of the site close to the 20-position of
the 1,4-substituted phenyl ring. This offered the possibility of
obtaining favorable or at least tolerated interactions with these
water molecules by the introduction of heteroatoms.
With this intention, a nitrogen was introduced into the
20-position of the 1,4-substituted phenyl ring. The resulting
compound 11 (Figure 4) displayed a similar p38R enzyme pot-
ency (pKi 7.6, n = 9) to the phenyl analogue 1 (pKi 7.9) and had
a moderate potency in the HWB assay (pIC50 6.5, n = 3).
A crystal structure of 11 bound to p38R (Figure 3) clearly shows
the interaction of the pyridine with the conserved water network.
Compared with 3, Tyr35 (not visible in Figure 3) has moved
from a position packed across the front of the inhibitor to allow
the pyridine nitrogen access to the water network, however this
change appears to have little effect on p38R potency.
Compound 11 has an excellent rat DMPK profile, with oral
bioavailability of 99%, low clearance (1.5 mL/mg/kg), and a
long half-life (iv T1/2 3.0 h); additionally, the cytochrome p450
profile was improved over its phenyl analogue (2C9 pIC50 4.4,
1A2, 2C19, 2D6, 3A4 pIC50<4). The isomeric pyridine with the
nitrogen inserted into the 30-position of 12 was prepared for
comparison, and 12 shows substantially lower p38R potency
(pKi 6.1, n=3), which is surprising given that the pyridine in
this position might be expected to help to rigidify the structure
by forming an internal hydrogen bond to the NH of the amide.
The 5-fluorine used to enhance HWB potency in the biphenyl
series was introduced into the 20-pyridyl series. The resulting
compound, 13, combines HWB potency (pIC50 7.1, n=11) and
a clean cytochrome p450 profile (pIC50<4.2 at all isoforms).
p38R and p38β were inhibited with pKi’s of 8.1 and 7.6,
respectively, and at least 100-fold selectivity was maintained
a nd, not determined.
against a panel of 33 kinases, including AKT1, ALK5, CDK2,
EGFR, ErbB2, GSK3β, IKK2, IKK-ε, JNK3, LCK,
PDGFR1B, PLK1, ROCK1, SRC, TIE2, and VEGFR2.
The excellent oral bioavailability seen for 11 was maintained
in the 5-Fanalogue (F>95%, Clr 2.5 mL/min/kg, iv T1/2 1.9 h).
Reoptimization of the cyclopropylmethylamide moiety of
13, by preparation of ca. 180 amides chosen as close analogues
and also by the virtual library process described earlier,
identified several further compounds that have increased
potency and good cytochrome p450 profiles. The most inter-
esting of these were ranked by oral AUC in rat cassette
DMPK studies (Table 3). The additional R-methyl group on
the amine in 15, 16, and 18 had a dramatic effect on the
exposure, reducing the oral AUC from 360 ng h/mL for
3
isobutylamide14to8ng h/mL and50ng h/mL, respectively,
3
3
for the R-methylated enantiomers 15 and 16 (0.25 mg/kg
dose). No further in vivo DMPK studies were performed on
15and16, as the lower oralexposure was consistentwiththein
vitro clearance data generated in rat liver microsomes.
Compounds 13, 14, and 2 were profiled in the rat PG-PS reac-
tivation model of arthritis; compound 14 generated a poor dose
response curve and no ED50 was calculated, but 13 and 2 gave
ED50s of 0.1 mg/kg and 0.03 mg/kg, respectively (Figure 5). The
rationale for the poor dose response for 14 in the PG-PS model is
unclear, as the cassette DMPK profiles of 2 and 14 are very simi-
lar (2: Clr 9.2 mL/min/kg, iv T1/2 1.1 h, F 51%; 14: Clr 6.4 mL/
min/kg, iv T1/2 1.1 h, F 56%. Dosed at 0.25 mg/kg po and iv).
Compound 2 also demonstrated an analgesic effect in the
rat PG-PS model, shown by increasing weight bearing on the

6260 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20
Aston et al.
Figure 7. Inhibition of clinical score in the murine CIA model
following 14 days of dosing with 2 (po bid), compared with
prednisolone (3 mg/kg po once daily). * = p<0.05 (Anova with
posthoc planned comparisons, treatment vs control).
inhibition of these kinases including AKT1, ALK5, CDK2,
EGFR, ErbB2, GSK3β, IKK2, IKK-ε, JNK3, LCK,
PDGFR1B, PLK1, ROCK1, SRC, TIE2, and VEGFR2 was
seen, and 2 has been shown to be highly selective for p38R/β.
Compound 2 has an excellent p450 profile (pIC50 2C9 4.5, 1A2,
2C19, 2D6, 3A4 e4.1) and moderate binding to rat and human
serum albumen (96.0 and 94.9%, respectively). Compound 2has
acceptable pharmacokinetics in four species, rat (discrete data at
1 mg/kg po and iv: F 71%, iv T1/2 1.2 h, Clr 8.1 mL/mg/kg. and
Vdss 0.9 mg/L), mouse (F 59%), dog (F 55%), and cynomolgus
monkey (F 35%), with a desirably low level of brain exposure
seen in the rat studies (brain:plasma ratio 0.1).
Figure 5. Effect of compounds 2 and 13 dosed orally twice daily on
inflammation in the PG-PS reactivation arthritis model assessed by
reduction in measured ankle swelling (mm). A comparison with
prednisolone dosed at 2.0 mg/kg once daily is shown. * = p < 0.05
(Anova with posthoc planned comparisons, treatment vs control).
** = p<0.01 (Anova with posthoc planned comparisons, treatment
vs control).
Conclusion
Using a combination of crystal structures of the bipheny-
lamides bound to p38R, classical medicinal chemistry, array,
and virtual array techniques, the BPA series has been opti-
mized from an initial low potency cross-screening hit,10 via a
compound with high enzyme potency but suboptimal proper-
ties 1, to a molecule suitable for clinical development 2. This
molecule has been developed for treatment of rheumatoid
arthritis and is also currently being clinically evaluated for
indications including COPD and heart disease
Chemistry. One of the advantages of the biphenylamide
series of p38R inhibitors is their straightforward and flexible
synthesis, either of the amide groups may be inserted in the
final step or both amides may be prepared, and the biphenyl
formed by a final Suzuki reaction (Scheme 1).
Figure 6. Effect of compound 2 on pain in the PG-PS reactivation
arthritis model assessed by weight bearing.
Compound 1 has been synthesized by a number of routes,
one of which (route B) has previously been described.12
Compounds 3-6 were prepared by route C, requiring the
trisubstituted benzoic acids 19a-d (Scheme 2). These were
not available commercially, so they were synthesized by
bromination of the disubstituted acids using bromine and
iron powder. While this preparation is useable on a small
scale, where there was a moderate and controllable
exotherm, scale-up of 19a required cooling due to a rapid
temperature rise partway through the portionwise addition
of the acid to the iron/bromine mixture. The resulting
product contained two regioisomers, as had been seen in
the small scale reaction, and also dibrominated material.
The 87 array compounds identified during the virtual
array process were prepared by either route C, 7, or by route
A, which allowed facile variation of the 40-amide, 8-10
(Scheme 3). To facilitate the larger arrays of amides, we
wished to develop an improved synthesis of 20a or preferably
injected paw on days 1 through 3 of the study, giving results
equivalent to prednisolone (2 mg/kg) at a dose of 0.3 mg/kg
(Figure 6).
In the murine CIA model, 2 showed dose related anti-
inflammatory and joint protective activity, reducing clinical
score, cellular influx into the synovial tissue and joint space,
cartilage destruction, and new bone formation. At 20 mg/kg
po bid, there was complete suppression of arthritis (Figure 7).
Further studies on compound 2 confirmed that, as well as
being competitive with the fluorescent ATP-site ligand used to
generate the SAR reported here, 2 was competitive with ATP
in an assay measuring the catalytic activity of activated p38R
with a Ki of 7.2. Compound 2 was profiled against a panel of 67
human kinases, using both in-house (pIC50 and % inhibition)
and Dundee panel (% inhibition) screening. No significant

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20 6261
Scheme 1. Biphenylamide Synthesis
Scheme 2. Synthetic Route to Compounds 3-6
Reagents and conditions: (i) Br2, Fe powder, rt (exothermic); (ii) SOCl2, reflux; (iii) cyclopropylamine, Na2CO3, DCM, rt; (iv) cyclopropylmethy-
lamine, Na2CO3, DCM, rt; (v) bispinnacolatodiboran, Pd(dppf)Cl2, KOAc, DMF, 80 °C; (vi) (Ph3P)4Pd, 1 M NaHCO3, IPA, 80 °C.
Scheme 3. 4-Amide Array Chemistry
Reagents and conditions: (i) NIS, trifluoromethylsulfonic acid, -20 °C; (ii) SOCl2 then cyclopropylamine, Na2CO3, DCM; (iii) {4-[(methyloxy)-
carbonyl]phenyl}boronic acid, (Ph3P)4Pd, 1 M NaHCO3, IPA, 80 °C; (iv) 2 M NaOH, MeOH; (v) RNH2, HATU, DIPEA, DMF/DCM, rt.

6262 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20
Aston et al.
Scheme 4. Synthesis of the Fluorinated Boronate Ester/Boronic
Acid
B at a flow rate of 3 mL/min. The mass spectra were recorded on a
Waters ZQ mass spectrometer using electrospray positive and
negative mode (ESþve and ES-ve). Unless otherwise specified,
1H NMR spectra were recorded using a 400 MHz spectrometer
and chemical shifts are given in ppm (δ) relative to tetramethylsi-
lane (TMS) as an internal standard. Purity of the tested com-
pounds was of >95% using interpretation of a combination of
LCMS and NMR data unless stated otherwise.
N -Cyclopropyl-N -(cyclopropylmethyl)-5-fluoro-6-methyl-3,40-
biphenyldicarboxamide (3). 3-Bromo-N-cyclopropyl-5-fluoro-4-
methylbenzamide (20a, 90 mg, 0.34 mmol), N-cyclopropylmethyl-
4-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)benzamide (22,
95 mg, 0.32 mmol), and tetrakis(triphenylphosphine)palladium
(6.0 mg, 5.2 ꢀ 10-3 mmol) and aq sodium hydrogen carbonate
(1M, 1.5 mL) were mixed in propan-2-ol (6 mL). The mixture was
heated at 90 °C for 18 h. The cooled reaction was adsorbed onto
silica under vacuum, and the residue was purified by chromatog-
raphy on a silica cartridge (10 g) eluting with an ethyl acetate/
cyclohexane gradient (0-75%). Appropriate fractions were re-
duced to dryness under vacuum and the residue triturated with
diethyl ether to give the desired product (70 mg, 60%) as a white
solid. 1H NMR (DMSO-d6) δ 8.65 (t, J = 5.5 Hz, 1H), 8.53 (d, J =
4.0 Hz, 1H), 7.97 (d, J = 8.0 Hz, 2H), 7.64 (d, J=11.0 Hz, 1H), 7.60
(1H, s), 7.49 (d, J = 8.0 Hz, 2H), 3.18 (t, J = 6.0 Hz, 2H), 2.85 (1H,
m), 2.17 (d, J = 2.0 Hz, 3H), 1.05 (1H, m), 0.70 (2H, m), 0.57 (2H,
m), 0.45 (2H, m), 0.25 (2H, m). LCMS: MHþ 367, retention time
3
4
0
Reagents and conditions: (i) bispinnacolatodiboran, Pd(dppf)Cl2,
KOAc, DMF, 80 °C; (ii) NaH, n-BuLi, (iPrO)3B, THF, -75 °C.
Scheme 5. Preparation of Pyridine Analogues
3.06 min. Anal. Calcd for C22H23FN2O2 0.4H2O: C, 70.72; H, 6.42;
3
N, 7.50. Found: C, 70.85; H, 6.19; N, 7.25.
3-Bromo-N-cyclopropyl-5-fluoro-4-methylbenzamide (20a).
3-Fluoro-4-methylbenzoic acid (460 mg, 3.0 mmol) was added to
a stirred mixture of bromine (2.3 mL, 45 mmol) and iron powder
(250 mg, 4.5 mmol) under nitrogen. The reaction was stirred at
20 °C for 4 h and then left to stand for 16 h. Sodium thiosulfate
solution (200 mL) was added, and the product was extracted into
ethyl acetate (3 ꢀ 150 mL). Ethyl acetate extracts were combined
and evaporated under vacuum. The crude product (19a, mixture
of isomers) was dissolved in DMF (7 mL). Cyclopropylamine
(0.21 mL, 3.0 mmol), HOBT (410 mg, 3.0 mmol), 1-(3-dimethyl-
aminopropyl)-3-ethylcarbodiimide hydrochloride (580 mg,
3.0 mmol), and DIPEA (530 μL, 3.0 mmol) were added to the
stirred solution. The reaction was stirred for 5 h at 20 °C. The
solvent was evaporated under vacuum, and the residue was parti-
tioned between ethyl acetate and water. The combined ethyl acetate
extracts were washed sequentially with aq sodium hydrogen carbo-
nate and hydrochloric acid (0.5M) and then dried (magnesium
sulfate). The ethyl acetate was evaporated under vacuum, and the
residue was purified by chromatography on a silica cartridge eluting
with cyclohexane/ethyl acetate (6:1) to give the desired product
Reagents and conditions: (i) SOCl2 80 °C; (ii) cyclopropylmethyla-
mine, Na2CO3, DCM rt; (iii) 25b, (Ph3P)4Pd, 1 M NaHCO3, DMF, 90 °C.
the equivalent iodide 23a, giving greater control of the
halogenation reaction and improved regioselectivity. The
bromination with iron and bromine was repeated at lower
temperatures (0-30 °C), but these conditions gave little
reaction. Various iodination conditions were assessed;
iodine with concentrated nitric and sulfuric acid gave a
controllable reaction but produced a mixture of isomers,
and greatest regioselectivity was found with n-iodosuccini-
mide in trifluoromethylsulfonic acid. On a small scale, this
chemistry was performed at 0 °C to give only the desired
isomer and uniodinated material, but for generating larger
quantities of material, reaction at -20 °C was found to give
improved control of regioselectivity.
To avoid the toxicity issues associated with stannane
chemistry for introduction of the 20-pyridine, 30-pyridine,
and 2,6-pyrimidine, we desired the boronate esters 25a-b or
the equivalent boronic acid 26a. These were readily prepared
from the iodides, either by palladium chemistry or by lithia-
tion and reaction with triisopropyl borate (Scheme 4).
The pyridines were then, simply prepared by Suzuki
reaction of the boronate with an appropriate halopyridine
(Scheme 5).
1
(360 mg, 44%). H NMR (CDCl3) δ 7.68 (s, 1H,), 7.39 (d, J =
9.5 Hz, 1H), 6.19 (bs, 1H), 2.88 (m, 1H), 2.36 (d, J = 2.5 Hz, 3H),
0.88 (m, 2H), 0.63 (m, 2H). LCMS: MHþ 272/274, retention time
3.12 min. [When this reaction was repeated on 10 times the scale,
with portionwise addition of the 3-fluoro-4-methylbenzoic acid
(4.6 g, 30 mmol) to the iron/bromine mixture, ice cooling of the
reaction was required part way through the addition due to a rapid
exotherm.]
N-Cyclopropylmethyl-4-(4,4,5,5-tetramethyl-[1,3,2]dioxaboro-
lan-2-yl)benzamide (22). N-(Cyclopropylmethyl)-4-iodobenza-
mide (21, 800 mg, 2.7 mmol), bispinnacolatodiboron (740 mg,
2.9 mmol), potassium acetate (1.2 g, 12 mmol), and Pd(dppf)Cl2
(50 mg, 0.07 mmol) were mixed in DMF (10 mL) and heated at
80 °C overnight. The cooled reaction was adsorbed onto silica
and purified by chromatography on a silica cartridge (10 g),
eluting with an ethyl acetate/cyclohexane gradient (0-100%).
Theappropriatefractions were reducedto dryness undervacuum
and the residue triturated with diethyl ether to give the desired
product as a white solid (550 mg, 69%). 1H NMR (DMSO-d6)
δ 8.63 (t, J = 5.7 Hz, 1H), 7.85 (d, J = 8.5 Hz, 2H), 7.73 (d, J =
8.0 Hz, 2H), 3.13 (t, J = 6.3 Hz, 2H), 1.30 (s, 12H), 1.04 (m, 1H),
0.41 (m, 2H), 0.22 (m, 2H).
Pyridines 2 and 13-18 were prepared in a similar fashion
from 25a or 26a and the appropriate chloronicotinamide.
Experimental Section
General Experimental Details. All commercial reagents and
solvents were obtained from commercial sources and used with-
out further purification. Unless otherwise specified, all reactions
were performed under nitrogen. LCMS was conducted on a
Supelcosil LCABZþPLUS column (3.3 cm ꢀ 4.6 mm ID) eluting
with 0.1% HCO2H and 0.01 M ammonium acetate in water
(solvent A) and 0.05% HCO2H 5% water in acetonitrile (solvent
B), using the following elution gradient 0.0-0.7 0% B, 0.7-
4.2 min 0-100% B, 4.2-4.6 min 100% B, 4.6-4.8 min 100-0%

Article
N-(Cyclopropylmethyl)-4-iodobenzamide (21). 4-Iodobenzoic
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20 6263
ylbenzamide (20c, 30 mg, 0.11 mmol), N-cyclopropylmethyl-
4-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)benzamide (22,
28 mg, 0.09 mmol), tetrakis(triphenylphosphine)palladium
(1.0 mg, 0.87 ꢀ 10-3 mmol), and aq sodium hydrogen carbonate
(1M, 0.5 mL) were mixed in propan-2-ol (2 mL) and heated at
90 °C under nitrogen for 24 h. The reaction was adsorbed onto
silica under vacuum, and the residue was purified by chroma-
tography on a silica cartridge (10 g) and eluted with an ethyl
acetate/cyclohexane gradient (0-100% ethyl acetate). The pro-
duct fractions were reduced to dryness under vacuum, and the
residue was recrystallized from ethyl acetate to give the desired
acid (1.0 g, 4.0 mmol) in thionyl chloride (3.0 mL, 41 mmol) was
heated at 100 °C for 30 min. The excess thionyl chloride was
evaporated under vacuum and the residue dissolved in DCM
(10 mL). Cyclopropylmethylamine (0.35 mL, 4.0 mmol) and
sodium carbonate (2.0 g, 19 mmol) were added to the solution
and the reaction left at ambient temperature for 2 h. The reaction
was filtered, the residue washed with DCM, and the combined
filtrate and washings reduced to dryness under vacuum. The
resulting material was triturated with diethyl ether to give the
desired product as a white solid (850 mg, 71%). 1HNMR(DMSO-
d6) δ 8.61 (t, J = 5.5 Hz, 1H), 7.84 (d, J = 8.5 Hz, 2H), 7.62 (d, J =
8.5 Hz, 2H), 3.11 (t, J = 6.5 Hz, 2H), 1.00 (m, 1H), 0.41 (m, 2H),
0.21 (m, 2H). LCMS: MHþ 302, retention time 2.98 min.
1
product as a white solid. H NMR (DMSO-d6) δ 8.62 (t, J =
5.5 Hz, 1H), 8.38 (d, J = 4.0 Hz, 1H), 7.94 (d, J = 8.5 Hz, 2H),
7.67 (s, 1H), 7.52 (s, 1H), 7.41 (d, J = 8.5 Hz, 2H), 3.18 (t, J =
6.0 Hz, 2H), 2.84 (m, 1H), 2.34 (s, 3H), 2.14 (s, 3H), 1.05 (m, 1H),
0.68 (m, 2H), 0.56 (m, 2H), 0.45 (m, 2H), 0.25 (m, 2H). LCMS:
MHþ 363, retention time 3.06 min.
3
4
0
5-Chloro-N -cyclopropyl-N -(cyclopropylmethyl)-6-methyl-
3,40-biphenyldicarboxamide (4). 3-Bromo-5-chloro-N-cyclopro-
pyl-4-methylbenzamide (20b, 30 mg, 2:1 mixture with 3-chloro-
N-cyclopropyl-4-methylbenzamide), N-cyclopropylmethyl-4-
(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)benzamide (22, 28 mg,
0.09 mmol), tetrakis(triphenylphosphine)palladium (1.0 mg, 0.87 ꢀ
10-3 mmol), and aq sodium hydrogen carbonate (1M, 0.5 mL) were
mixed in propan-2-ol (2 mL) and heated at 90 °C under nitrogen for
24 h. The reaction was adsorbed onto silica under vacuum, the residue
purified by chromatography on a silica cartridge (5 g) eluting with an
ethyl acetate/cyclohexane gradient (0-100%). The product fractions
were reduced to dryness under vacuum and the residue recrystallized
3-Bromo-N-cyclopropyl-4,5-dimethylbenzamide (20c). 3-Bro-
mo-4,5-dimethylbenzoic acid23 (19c, 200 mg, 0.87 mmol) was
mixed with thionyl chloride (2.0 mL, 27 mmol) and the mixture
heated at 90 °C for 2.5 h. The excess thionyl chloride was
evaporated under vacuum, and the residue was dissolved in
DCM (5 mL). Cyclopropylamine (0.2 mL, 2.9 mmol) and
sodium carbonate (∼300 mg) were added to the solution and
the reaction stirred for 2 h at room temperature. The reaction
was filtered, the filtrate reduced to dryness under vacuum, and
the residue triturated with diethyl ether. The resulting solid was
dissolved in acetone/methanol and adsorbed onto silica under
vacuum. The residue was purified by chromatography on a silica
cartridge (5 g) eluting with an ethyl acetate/cyclohexane gradi-
ent (0-50% ethyl acetate) to give the desired product. 1H NMR
(DMSO-d6) δ 8.44 (d, J = 4.0 Hz, 1H), 7.87 (s, 1H), 7.64 (s, 1H),
2.83 (m, 1H), 2.34 (s, 6H), 0.69 (m, 2H), 0.57 (m, 2H). LCMS:
MHþ 268/270, retention time 3.05 min.
1
from ethyl acetate to give the desired product as a white solid. H
NMR (DMSO-d6) δ 8.65 (t, J = 5.5 Hz), 1H), 8.57 (d, J = 4.5 Hz,
1H), 7.96 (d, J = 8.5 Hz, 2H), 7.93 (d, J = 1.5 Hz, 1H), 7.68 (d, J =
1.5 Hz, 1H), 7.48 (d, J = 8.5 Hz, 2H), 3.18 (t, J = 6.0 Hz, 2H), 2.86
(m, 1H), 2.27 (s, 3H), 1.05 (m, 1H), 0.69 (m, 2H), 0.57 (m, 2H), 0.45
(m, 2H), 0.25 (m, 2H). LCMS: MHþ 383/385, retention time
3.27 min.
3-Bromo-5-chloro-N-cyclopropyl-4-methylbenzamide (20b).
3-Bromo-5-chloro-4-methylbenzoic acid (19b, 310 mg, 2:1 mix-
ture with 3-chloro-4-methylbenzoic acid) was mixed with thionyl
chloride (3.0 mL, 41 mmol) and the mixture heated at 90 °C for
2.5 h. The excess thionyl chloride was evaporated under vacuum,
and the residue was dissolved in DCM (7.5 mL). Cyclopropyla-
mine (0.20 mL, 2.9 mmol) and sodium carbonate (∼500 mg) were
added to the solution and the reaction stirred for 2 h at room
temperature. The reaction was filtered, the residue washed with
DCM, and the combined filtrate and washings were reduced to
dryness under vacuum and the resulting solid triturated with
cyclohexane. The resulting white solid was adsorbed onto silica
under vacuum and purified by chromatography on a silica
cartridge (10 g), eluting with an ethyl acetate/cyclohexane gra-
dient (0-50%). The product fractions were reduced to dryness
under vacuum to give a mixture of 3-bromo-5-chloro-N-cyclo-
propyl-4-methylbenzamide and 3-chloro-N-cyclopropyl-4-meth-
ylbenzamide (2:1). 3-Bromo-5-chloro-N-cyclopropyl-4-methyl-
3
4
0
N -cyclopropyl-N -(cyclopropylmethyl)-4-fluoro-6-methyl-3,40-
biphenyldicarboxamide (6). 5-Bromo-N-cyclopropyl-2-fluoro-4-
methylbenzamide (20d, 30 mg, 0.11 mmol), N-cyclopropylmethyl-
4-(4,4,5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)benzamide (22, 28
mg, 0.09 mmol), tetrakis(triphenylphosphine)palladium (1.0 mg,
0.87 ꢀ 10-3 mmol), and aq sodium hydrogen carbonate (1M,
0.5 mL) were mixed in propan-2-ol (2 mL) and heated at 90 °C
under nitrogen for 24 h. The reaction was adsorbed onto silica under
vacuum, and the residue was purified by chromatography on a silica
cartridge (5 g) eluting with an ethyl acetate/cyclohexane gradient
(0-100% ethyl acetate). The product fractions were reduced to
dryness under vacuum and the residue recrystallized from ethyl
acetate to give the desired product as a white solid. 1H NMR
(DMSO-d6) δ8.63 (t, J= 6.0 Hz, 1H), 8.35 (d, J= 4.0 Hz, 1H), 7.93
(d, J = 8.0 Hz, 2H), 7.45 (d, J = 8.5 Hz, 2H), 7.39 (d, J = 7.5 Hz,
1H), 7.27 (d, J = 11.5 Hz, 1H), 3.17 (t, J = 6.5 Hz, 2H), 2.83 (m,
1H), 2.26 (s, 3H), 1.05 (m, 1H), 0.69 (m, 2H), 0.54 (m, 2H), 0.45 (m,
2H), 0.25 (m, 2H). LCMS: MHþ 367, retention time 2.97 min.
5-Bromo-N-cyclopropyl-2-fluoro-4-methylbenzamide (20d).
5-Bromo-2-fluoro-4-methylbenzoic acid (19d, 180 mg, 0.77 mmol)
was mixed with thionyl chloride (2.0 mL, 27 mmol) and the mixture
heated at 90 °C for 2 h. The excess thionyl chloride was evaporated
under vacuum, and the residue was dissolved in DCM (5 mL).
Cyclopropylamine (0.10 mL, 1.4 mmol) and sodium carbonate
(∼300 mg) were added to the solution and the reaction stirred for
2 h at room temperature. The reaction was filtered, the residue
washed with DCM, and the combined filtrate and washings
reduced to dryness under vacuum. The residue was recrystallized
from cyclohexane to give the desired product as a white solid. 1H
NMR (DMSO-d6) δ 8.41 (s, 1H), 7.71 (d, J = 6.8 Hz, 1H), 7.36
(d, J = 10.8 Hz, 1H), 2.81 (m, 1H), 2.36 (s, 3H), 0.69 (m, 2H), 0.55
(m, 2H). LCMS: MHþ 272/274, retention time 2.88 min.
1
benzamide H NMR (DMSO-d6) δ 8.61 (d, J = 3.5 Hz 1H),
8.03 (s, 1H), 7.90 (s, 1H), 2.85 (m, 1H), 2.49 (s, 3H), 0.70 (m, 2H),
0.58 (m, 2H). LCMS: MHþ 288/290/292, retention time 3.28 min.
3-Bromo-5-chloro-4-methylbenzoic acid (19b). 3-Chloro-4-
methylbenzoic acid (270 mg, 1.6 mmol) was added in portions
to a mixture of bromine (1.0 mL, 20 mmol) and iron powder
(45 mg, 0.81 mmol) and the reaction stirred in a sealed vial for
28 h. The reaction mixture was poured into aq sodium thiosul-
fate and extracted with ethyl acetate. The combined extracts
were washed with brine, dried (magnesium sulfate), and the
solvent evaporated under vacuum. The residue was redissolved
in ethyl acetate, filtered, and the filtrate reduced to dryness
under vacuum to give a mixture of 3-bromo-5-chloro-4-methyl-
benzoic acid and 3-chloro-4-methylbenzoic acid (320 mg, 2:1).
3-Bromo-5-chloro-4-methylbenzoic acid 1H NMR (DMSO-d6)
δ 8.02 (d, J = 1.5 Hz, 1H), 7.89 (d, J = 1.5 Hz, 1H), 2.51 (s, 3H).
LCMS: [M - H]- 247/249/251, retention time 3.73 min.
5-Bromo-2-fluoro-4-methylbenzoic Acid (19d). 2-Fluoro-4-
methylbenzoic acid24 (240 mg, 1.6 mmol) was added in portions
to a mixture of bromine (1.0 mL, 20 mmol) and iron powder
(60 mg, 1.1 mmol) and the reaction stirred in a sealed reacti-vial
3
4
0
N -Cyclopropyl-N -(cyclopropylmethyl)-5,6-dimethyl-3,40-bi-
phenyldicarboxamide (5). 3-Bromo-N-cyclopropyl-4,5-dimeth-

6264 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20
Aston et al.
at room temperature for 25 min. The reaction was poured into
aq sodium thiosulfate and extracted with ethyl acetate (ꢀ2). The
combined extracts were washed with brine, dried (magnesium
sulfate), and the solvent evaporated under vacuum. The residue
was recrystallized from cyclohexane to give the desired product
with methanol (35 mL) and sodium hydroxide (5 mL) at 50 °C
for 2 h. The mixture was reduced to dryness under vacuum and
the residue dissolved in ethyl acetate water (1:1, 60 mL). The
solution was acidified to pH2 with hydrochloric acid (2M), the
layers separated, and the aqueous extracted with ethyl acetate
(30 mL). The combined organic phases were washed with water,
dried (magnesium sulfate), and reduced to dryness under va-
cuum. The residue was dissolved in ethyl acetate (30 mL),
extracted with sodium hydroxide (2M, 2 ꢀ 50 mL), and the
aqueous phase acidified to pH 1 and extracted with ethyl acetate
(3 ꢀ 30 mL). The combined organics were dried (magnesium
sulfate) and reduced to dryness under vacuum. The resulting
solid was dried under vacuum for 2 h to give the desired product
(0.28 g, 34%). 1H NMR (MeOD) δ 8.11 (d, J = 8.3 Hz, 2H), 7.55
(m, 2H), 7.46 (d, J = 8.0 Hz, 2H), 2.84 (m, 1H), 2.20 (s, 3H), 0.78
(m, 2H), 0.62 (m, 2H). LCMS: MHþ 314, retention time 3.08 min.
N-Cyclopropyl-3-fluoro-5-iodo-4-methylbenzamide (23a). 3-
Fluoro-5-iodo-4-methylbenzoic acid (29, 220 g, 0.77 mol) in
thionyl chloride (270 mL, 3.70 mol) was heated at 100 °C for 3 h.
The excess thionyl chloride was evaporated under vacuum and
the residue azeotroped with DCM (ꢀ2). The resulting acid
chloride was dissolved in DCM (200 mL) and the solution added
slowly to a cold (ice/water bath), stirred suspension of cyclo-
propylamine (110 mL, 1.6 mol) and sodium carbonate (220 g,
2.0 mol) in DCM (500 mL). The reaction was stirred overnight
at ambient temperature, diluted with DCM, and washed with
water (ꢀ2). The combined aqueous washings were extracted
with DCM, the DCM fractions combined, dried (sodium
sulfate), and partially concentrated under vacuum. The result-
ing suspension was allowed to stand for 3 h and the solid isolated
by filtration, washed with DCM, and dried to give the desired
product (184 g). A further batch of the desired product (11 g)
was obtained by evaporation of the solvent from the mother
liquor and recrystallization of the residue from DCM (150 mL).
A final batch of the desired product (30 g) was prepared
by chromatography of the mother liquor on a silica cartridge
(800 g) eluting with DCM and then ethyl acetate/DCM (5:95).
Total yield (225 g, 92%). 1H NMR (600 M Hz, CDCl3) δ 7.93
(s, 1H), 7.42 (d, J = 9.6 Hz, 1H), 6.42 (bs, 1H), 2.88 (m, 1H), 2.38
(s, 3H), 0.87 (m, 2H), 0.64 (m, 2H). LCMS: MHþ 320, retention
time 3.23 min.
3-Fluoro-5-iodo-4-methylbenzoic acid (29). 3-Fluoro-4-methyl-
benzoic acid (180 g, 1.2 mol) in trifluoromethanesulfonic acid
(1.1 L) was cooled to -20 °C. n-Iodosuccimide (270 g, 1.2 mol)
was added in portions over 75 min maintaining a temperature of
-18 to -19 °C. The reaction was stirred at -20 °C for 4 h, further
NIS (55 g, 0.24 mol) added, and the reaction stirred overnight at
-20 °C. NIS (19 g, 84 mmol) was added to the reaction and
stirring continued for 24 h. The reaction was warmed up to -5 °C
and the suspension poured into ice (3 kg) and 10% sodium
thiosulfate solution. The aqueous suspension was filtered, and
the solid was washed with water and partially dried. The solid was
partitioned between ethyl acetate (5 L) and 10% sodium thiosul-
fate solution (1.5 L) and the organic phase washed with sodium
thiosulfate solution. The combined aqueous phases were extrac-
ted with ethyl acetate (2 L); the organic phases were combined,
dried (sodium sulfate), and concentrated under vacuum to ca.
600 mL. The slurry was allowed to stand for 4 h, the solid filtered
off, washed with ethyl acetate, and dried to give the desired
product (215 g, 65%). 1H NMR (250 M Hz, CDCl3) δ 10.76 (bs,
1H), 8.34 (s, 1H), 7.72 (d, 1H), 2.45 (s, 3H). LCMS: [M - H]-
279, retention time 3.75 min.
1
as a white solid (180 mg, 48%). H NMR (DMSO-d6) δ 13.44
(bs, 1H), 7.98 (d, J = 7.0 Hz, 1H), 7.41 (d, J = 11.5 Hz, 1H), 2.39
(s, 3H). LCMS: [M - H]- 231/233, retention time 3.37 min.
3
4
0
N -Cyclopropyl-5-fluoro-6-methyl-N -(tetrahydro-2-furanyl-
methyl)-3,40-biphenyldicarboxamide (7). N-Cyclopropyl-3-fluoro-
4-methyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzamide
(25a, 30 mg, 0.10 mmol), 4-bromo-N-(tetrahydro-2-furanylmethyl)-
benzamide (28, 27 mg, 0.10 mmol), tetrakis(triphenylphosphine)-
palladium (7.0 mg, 6.1 ꢀ 10-3 mmol), and aq sodium hydrogen
carbonate (1M, 0.45 mL) were mixed in propan-2-ol (2 mL)
and heated at 90 °C under nitrogen for 18 h. The reaction was
filtered, the residue washed with DCM, and the combined
filtrate and washings reduced to dryness. The resulting materi-
al was purified by chromatography on a silica cartridge (8 g)
eluting with ethyl acetate/cyclohexane (1:2, then 1:1 and 3:1).
Appropriate fractions were reduced to dryness under vacuum
to give the desired product (17 mg, 45%). 1H NMR (DMSO-d6)
δ 8.64 (t, J = 5.5 Hz, 1H), 8.53 (d, J = 4.0 Hz, 1H), 7.96 (d, J =
8.5 Hz, 2H), 7.63 (d, J = 10.5 Hz, 1H), 7.60 (s, 1H), 7.49 (d, J =
8.5 Hz, 2H), 4.00 (m, 1H), 3.79 (q, J = 7.0 Hz, 1H), 3.64 (q, J =
7.0 Hz, 1H), 3.36 (m partially obscured by water, 2H), 2.85 (m,
1H), 2.17 (s, 3H), 1.95-1.77 (m, 3H), 1.63 (m, 1H), 0.70 (m, 2H),
0.57 (m, 2H). LCMS: MHþ 396, retention time 3.01 min.
4-Bromo-N-(tetrahydro-2-furanylmethyl)benzamide (28). A
solution of 4-bromobenzoyl chloride (500 mg, 2.3 mmol) in
DCM (5 mL) was treated with sodium carbonate (290 mg) and
(tetrahydro-2-furanylmethyl)amine (230 mg, 2.3 mmol) and
the mixture stirred at ambient temperature under nitrogen for
16 h. The reaction was filtered, and the solid was washed with
DCM. The combined filtrate and washings were filtered
through an aminopropyl ion exchange cartridge and the fil-
trate evaporated under vacuum to give the desired product as a
white solid (280 mg, 42%). 1H NMR (DMSO-d6) δ 8.63 (t, 1H),
7.80 (d, J = 8.5 Hz, 2H), 7.67 (d, J = 8.5 Hz, 2H), 3.96 (m, 1H),
3.77 (q, J = 7.5 Hz, 1H), 3.62 (q, J = 7.4 Hz, 1H), 3.30 (m, 2H),
1.94-1.75 (m, 3H), 1.59 (m, 1H). LCMS: MHþ 284/286,
retention time 2.66 min. Purity 85%.
3
4
0
N -Cyclopropyl-5-fluoro-6-methyl-N -[(2-methyl-1,3-thiazol-
4-yl)methyl]-3,40-biphenyldicarboxamide (8).
solution of
A
50-[(cyclopropylamino)carbonyl]-30-fluoro-20-methyl-4-biphenyl-
carboxylic acid (24, 310 mg, 1.0 mmol) in DMF (4.2 mL) was
treated with HATU (400 mg, 1.05 mmol) and diisopropylethyla-
mine (450 μL, 2.0 mmol). After 10 min, 200 μL of this mixture was
added to 200 μL of a solution of [(2-methyl-1,3-thiazol-4-yl)-
methyl]amine (260 mg, 0.2 mmol) in DMF (2 mL). The reaction
was left at ambient temperature for 16 h before the solvents were
evaporated under vacuum. The residue was dissolved in chloro-
form, loaded onto an aminopropyl ion exchange cartridge (0.5 g),
the cartridge washed with chloroform, and the basic material
eluted with ethyl acetate/methanol (9:1). The combined chloro-
form fractions were reduced to dryness to give the desired
compound. 1H NMR (600 M Hz, DMSO-d6) δ 9.11 (t, J =
6.0 Hz, 1H), 8.52 (d, J = 4.0 Hz, 1H), 8.02 (d, J = 8.5 Hz, 2H),
7.63 (d, J = 10.0 Hz, 1H), 7.60 (s, 1H), 7.51 (d, J = 8.5 Hz, 2H),
7.22 (s, 1H), 4.55 (d, J = 6.0 Hz, 2H), 2.85 (m, 1H), 2.64 (s, 3H),
2.17 (s, 3H), 0.70 (m, 2H), 0.57 (m, 2H). LCMS: MHþ 424,
retention time 2.95 min.
3
4
50-[(Cyclopropylamino)carbonyl]-30-fluoro-20-methyl-4-biphe-
nylcarboxylic acid (24). A mixture of N-cyclopropyl-3-fluoro-
5-iodo-4-methylbenzamide (23a, 0.85 g, 2.7 mmol), {4-[(methyl-
oxy)carbonyl]phenyl}boronic acid (0.48 g, 2.7 mmol), tetrakis-
(triphenylphosphine)palladium (0.31 g, 0.27 mmol), and aq
sodium hydrogen carbonate (1M, 13 mL) in propan-2-ol
(28 mL) was heated at 90 °C for 22 h under nitrogen. The
solvents were evaporated under vacuum and the residue treated
N -Cyclopropyl-5-fluoro-6-methyl-N -[2-(methyloxy)ethyl]-3,40-
biphenyldicarboxamide (9). A solution of 50-[(cyclopropylamino)-
carbonyl]-30-fluoro-20-methyl-4-biphenylcarboxylic acid (24, 0.31g,
1.0 mmol) in DMF (4.2 mL) was treated with HATU (0.40 g,
1.1 mmol) and diisopropylethylamine (450 μL, 2.0 mmol). After
10 min, 200 μL of this mixture was added to 200 μL of a solution of
2-methoxyethylamine (150 mg, 0.20 mmol) in DMF (2 mL). The
reaction was left at ambient temperature for 16 h before the solvents
0

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20 6265
were evaporated under vacuum. The residue was dissolved in
chloroform, loaded onto an aminopropyl ion exchange cartridge
(0.5 g), the cartridge washed with chloroform, and the basic material
eluted with ethyl acetate/methanol (9:1). The combined chloroform
fractions were reduced to dryness to give the desired compound. 1H
NMR (DMSO-d6) δ 8.61 (t, J = 6.0 Hz, 1H), 8.52 (d, J = 4.0 Hz,
1H), 7.96 (d, J = 8.5 Hz, 2H), 7.63 (d, J = 11.0 Hz, 1H), 7.59 (s,
1H), 7.49 (d, J = 8.5 Hz, 2H), 3.49-3.44 (4H, m), 3.28 (s, 3H), 2.17
(d, J = 2.0 Hz, 3H), 0.70 (m, 2H), 0.57 (m, 2H). LCMS: MHþ 371,
retention time 2.81 min, purity 88%.
was triturated with diethyl ether to give N-cyclopropyl-3-
iodo-4-methylbenzamide as a white solid (1.1 g, 96%). 1H
NMR (DMSO-d6): δ 8.46 (d, J = 4.0 Hz, 1H), 8.24 (d, J =
1.5 Hz, 1H), 7.74 (dd, J = 7.8 Hz, J = 1.8 Hz, 1H), 7.38 (d, J =
8.0 Hz, 1H), 2.82 (m, 1H), 2.38 (s, 3H), 0.67 (m, 2H), 0.55
(m, 2H).
6-Chloro-N-cyclopropylmethylnicotinamide (27a). 6-Bromo-
nicotinic acid (1.5 g, 7.6 mmol) was heated at 100 °C in thionyl
chloride (5.0 mL, 69 mmol) for 2 h. The excess thionyl chloride
was evaporated under vacuum and residue was dissolved in
DCM (15 mL), and sodium carbonate (1.5 g) was added to the
mixture followed by slow addition of cyclopropylmethylamine
(2.0 mL). The reaction was stirred at room temperature over-
night, filtered, and the residue washed with DCM (30 mL) and
then ethyl acetate (30 mL). The combined filtrate and washings
were reduced to dryness under vacuum. The residue was purified
by dissolving in DCM, applying to a silica cartridge and eluting
with ethyl acetate. The solvent was evaporated from the eluent
fractions under vacuum and the residue further dried under
vacuum to give 6-chloro-N-cyclopropylmethylnicotinamide as
a white solid (1.2 g, 62%). 1H NMR (DMSO-d6) δ 8.83 (m, 2H),
8.25 (dd, J = 8.3 Hz, J = 2.5 Hz, 1H), 7.65 (d, J = 8.3 Hz, 1H),
3.15 (t, J = 5.8 Hz, 2H), 1.02 (m, 1H), 0.44 (m, 2H), 0.24
(m, 2H).
5-{5-[(Cyclopropylamino)carbonyl]-2-methylphenyl}-N-(cyclo-
propylmethyl)-2-pyridinecarboxamide (12). 5-Bromo-N-(cyclo-
propylmethyl)-2-pyridinecarboxamide (27b, 44 mg, 0.27 mmol)
and N-cyclopropyl-4-methyl-3-(4,4,5,5-tetramethyl-[1,3,2]dioxa-
borolan-2-yl)-benzamide (25b, 50 mg, 0.17 mmol), aq sodium
carbonate (2N, 1 mL), and tetrakis(triphenylphosphine)palladium
(6 mg, 5.2 ꢀ 10-3 mmol) were heated at 80 °C in DMF (3 mL) for
overnight. The reaction was diluted with ethyl acetate, dried
(magnesium sulfate), and the solution filtered through a silica
cartridge (5 g). The cartridge was washed with further ethyl acetate
and the combined filtrate and washings reduced to dryness under
vacuum. The residue was triturated with diethyl ether to give the
desired product. 1H NMR (DMSO-d6) δ 9.03 (d, J = 2.0 Hz, 1H),
8.82 (t, J = 5.5 Hz, 1H), 8.73 (d, J = 2.0 Hz, 1H), 8.44 (d, 4.5 Hz,
1H), 8.21 (d, J = 2.5 Hz, d), 7.80 (dd, J = 8.0 Hz, J = 1.5 Hz, 1H),
7.75 (d, J = 2.0 Hz, 1H), 7.43 (d, J = 8.0 Hz, 1H), 3.17 (t, J =
6.5 Hz, 2H), 2.84 (m, 1H), 2.27 (s, 3H), 1.05 (m, 1H), 0.67 (m, 2H),
0.55 (m, 2H), 0.44 (m, 2H), 0.23 (m, 2H). LCMS: MHþ 350,
retention time 2.62 min
5-Bromo-N-(cyclopropylmethyl)-2-pyridinecarboxamide (27b).
5-Bromo-2-pyridinecarboxylic acid (51 mg, 0.25 mmol) was
heated in thionyl chloride (1.0 mL, 14 mmol) at 85 °C for 2 h.
The excess thionyl chloride was evaporated under vacuum, and
the residue was dissolved in DCM (2 mL) and treated with
cyclopropylmethylamine (0.10 mL) and potassium carbonate
(200 mg). The reaction was stirred at room temperature for
∼2 h and then left to stand for 72 h. The reaction was filtered
and the residue washed with DCM (∼2 mL). The combined
filtrate and washings (pale-yellow solution) were reduced to
dryness under a stream of nitrogen. The residual gum was
triturated with diethyl ether, filtered, and the filtrate diluted with
cyclohexane. The solvents were evaporated under vacuum to give
the desired product as cream-colored waxy solid (44 mg, 68%).
1H NMR (DMSO-d6) δ 8.81 (bt, 1H), 8.77 (d, J = 2.3 Hz, 1H),
8.24 (dd, J = 8.3 Hz, J = 2.3 Hz, 1H), 7.97 (d, J = 8.3 Hz, 1H),
3.16 (t, J = 6.5 Hz, 2H), 1.05 (m, 1H), 0.42 (m, 2H), 0.24 (m, 2H).
LCMS: MHþ 255/257, retention time 2.83 min.
3
4
0
N -cyclopropyl-5-fluoro-N -(2-hydroxy-2-methylpropyl)-6-meth-
yl-3,40-biphenyldicarboxamide (10). 50-[(Cyclopropylamino)carbo-
nyl]-30-fluoro-20-methyl-4-biphenylcarboxylic acid (24, 50 mg, 0.16
mmol), HATU (61 mg, 0.16 mmol), 1-amino-2-methyl-2-propanol
(21 mg, 0.24 mmol), and diisopropylethylamine (31 μL, 0.18 mmol)
were mixed in DMF (2 mL) and stirred at ambient temperature
under nitrogen for 24 h. A further aliquot of 1-amino-2-methyl-2-
propanol (10 mg, 0.12 mmol) was added and stirring continued for
24 h. The reaction was reduced to dryness under vacuum and
purified by chromatography on a silica cartridge (10 g) eluting with
an ethyl acetate/cyclohexane gradient (50-100%) and then metha-
nol/ethyl acetate (1-3%). Appropriate fractions were reduced to
1
dryness, to give the desired compound (35 mg, 57%). H NMR
(DMSO-d6) δ 8.53 (d, J = 4.0 Hz, 1H), 8.34 (t, J = 6.0 Hz, 1H),
7.97 (d, J = 8.5 Hz, 2H), 7.63 (d, J = 10.5 Hz, 1H), 7.60 (s, 1H),
7.49 (d, J = 8.0 Hz, 2H), 4.58 (s, 1H), 3.29 (d, J = 6.0 Hz, 2H), 2.85
(m, 1H), 2.17 (d, J = 2.5 Hz, 3H), 1.12 (s, 3H), 0.70 (m, 2H), 0.57
(m, 2H). LCMS: MHþ 385, retention time 2.73 min.
6-(5-Cyclopropylcarbamoyl-2-methyl-phenyl)-N-cyclopropyl-
methyl-nicotinamide (11). 6-Chloro-N-cyclopropylmethylnicoti-
namide (27a, 26 mg, 0.10 mmol), N-cyclopropyl-4-methyl-3-(4,4,
5,5-tetramethyl-[1,3,2]dioxaborolan-2-yl)-benzamide (25b, 30 mg,
0.10 mmol), aq sodium carbonate (2N, 0.5 mL), and tetrakis-
(triphenylphosphine)palladium (4.0 mg, 3.5 x10-3 mmol) were
heated at 90 °C in DMF (1 mL) for 3 h. The reaction was adsorbed
onto silica and purified by chromatography on a silica cartridge
(5 g), eluting with an ethyl acetate/cyclohexane gradient (0 to
100%), and then acetone. The solvent was evaporated from the
product fractions under vacuum and the residue triturated with
diethyl ether to give the desired product as a cream solid. 1H NMR
(DMSO-d6) δ 9.11 (s, 1H), 8.84 (t, J = 5.5 Hz, 1H), 8.48 (d, J =
4.0 Hz, 1H), 8.31 (dd, J=8.0Hz, J= 2.0 Hz, 1H), 7.88 (s, 1H), 7.81
(d, J = 8.0 Hz, 1H), 7.70 (d, J = 8.5 Hz, 1H), 7.41 (d, J = 8.0 Hz,
1H), 3.20 (t, J = 6.5 Hz, 2H), 2.86 (m, 1H), 2.37 (s, 3H), 1.06 (m,
1H), 0.69 (m, 2H), 0.57 (m, 2H), 0.46(m, 2H), 0.26(m, 2H). LCMS:
MHþ 350, retention time 2.70 min
N-Cyclopropyl-4-methyl-3-(4,4,5,5-tetramethyl-[1,3,2]dioxa-
borolan-2-yl)-benzamide (25b). N-Cyclopropyl-3-iodo-4-meth-
ylbenzamide (23b, 1.1 g, 3.6 mmol), bis(pinnacolato)diboron
(1.9 g, 7.3 mmol), potassium acetate (1.8 g, 18 mmol), and
Pd(dppf)Cl2 (55 mg, 75ꢀ10-3 mmol) were heated at 85 °C in
DMF (30 mL) for 4.5 h. The cooled reaction was adsorbed onto
silica and purified by chromatography on a silica cartridge
(10 g), eluting with an ethyl acetate/cyclohexane gradient (0 to
100%). The solvent was evaporated from the product fractions
under vacuum and the residue triturated with cyclohexane to
1
give the desired product as a beige solid (650 mg, 59%). H
NMR (DMSO-d6) δ 8.40 (d, J = 4.0 Hz, 1H), 8.06 (d, J =
2.0 Hz, 1H), 7.76 (dd, J=8.0 Hz, J=2.0 Hz, 1H), 7.23 (d, J =
8.0 Hz, 1H), 2.82 (m, 1H), 2.48 (s, 3H), 1.30 (s, 12H), 0.66
(m, 2H), 0.56 (m, 2H).
N-Cyclopropyl-3-iodo-4-methylbenzamide (23b). 3-Iodo-4-
methylbenzoic acid (1.0 g, 3.8 mmol) was heated at 80 °C
in thionyl chloride (10 mL, 140 mmol) for 2 h. The reaction
was allowed to cool to room temperature and the excess thionyl
chloride evaporated under vacuum. The residue was dissolved
in DCM (10 mL), and cyclopropylamine (0.32 mL, 4.6 mmol)
and sodium carbonate (2.0 g) were added to the solution. The
reaction was stirred at room temperature for 18 h, filtered,
and the filtrate reduced to dryness under vacuum. The residue
6-{5-[(Cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}-
N-(cyclopropylmethyl)-3-pyridinecarboxamide (13). 6-Chloro-
N-(cyclopropylmethyl)-3-pyridinecarboxamide (27a, 26 mg,
0.10 mmol), N-cyclopropyl-5-fluoro-4-methyl-3-(4,4,5,5-tetra-
methyl-[1,3,2]dioxaborolan-2-yl)-benzamide (25a, 32 mg, 0.10
mmol), tetrakis(triphenylphosphino)palladium (2.0 mg, 1.7 ꢀ
10-3 mmol), and aq sodium hydrogen carbonate (1M, 0.5 mL)
were mixed in propan-2-ol (2 mL) and heated at reflux for 18 h.
The cooled reaction was diluted with ethyl acetate and applied to

6266 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20
Aston et al.
a silica cartridge (5 g), the cartridge was washed with ethyl
acetate and the combined filtrate and washings reduced to
dryness under vacuum. The residual glass was dissolved in ethyl
acetate/methanol and applied to an SCX-2 ion exchange car-
tridge. The product was eluted with 0.88 ammonia/ethyl acetate/
methanol, the solvents evaporated under vacuum, and the
residue triturated with diethyl ether/cyclohexane to give the
desired product as a white solid (18 mg, 49%). 1H NMR
(DMSO-d6) δ 9.14 (d, J=1.5 Hz, 1H), 8.87 (t, J=5.5 Hz, 1H),
8.58 (d, J = 3.5 Hz, 1H), 8.35 (dd, J = 8.5 Hz, J = 2.0 Hz, 1H),
7.78 (s, 1H), 7.75-7.69 (m, 2H), 3.21 (t, J = 6.0 Hz, 2H), 2.87
(m, 1H), 2.27 (s, 3H), 1.08 (m, 1H), 0.71 (m, 2H), 0.60 (m, 2H),
0.48 (m, 2H), 0.28 (m, 2H). LCMS: MHþ 368, retention time
2.81 min. Anal. Calcd. for C21H22FN2O2: C, 68.65; H, 6.04; N,
11.44. Found: C, 68.32; H, 6.01; N, 11.17.
N-Cyclopropyl-5-fluoro-4-methyl-3-(4,4,5,5-tetramethyl-[1,3,2]-
dioxaborolan-2-yl)-benzamide (25a). N-Cyclopropyl-3-fluoro-5-
iodo-4-methylbenzamide (23a, 6.0 g, 19 mmol), bispinnacolatodi-
boron (7.2 g, 28 mmol), potassium acetate (9.2 g, 94 mmol), and
Pd(dppf)Cl2 (150 mg, 0.20 mmol) were mixed in anhydrous DMF
(130 mL) and heated at 90 °C for 19 h. The cooled reaction was
adsorbed onto silica and purified by chromatography on silica
cartridges (40 g ꢀ 2) eluting with ethyl acetate/cyclohexane
(10 then 15%). The product fractions were combined the solvents
evaporated under vacuum. The residue was recrystallized from
cyclohexane to give the desired product as a white solid (3.0 g,
54%). 1H NMR (CDCl3) δ 7.73 (s, 1H), 7.60 (d, J = 11.8 Hz, 1H),
6.32 (bs, 1H), 2.89 (m, 1H), 2.49 (d, J = 2.3Hz, 3H), 1.36 (s, 12H),
0.88 (m, 2H), 0.65 (m, 2H).
6-{5-[(Cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}-
N-(2-methylpropyl)-3-pyridinecarboxamide (14). Compound 14
was prepared as for 2 using 6-chloro-N-(2-methylpropyl)-3-
pyridinecarboxamide (31). 1H NMR (DMSO-d6) δ 9.11 (d,
J = 1.5 Hz, 1H), 8.74 (t, J = 6.0 Hz, 1H), 8.56 (d, J = 4.0 Hz,
1H), 8.32 (dd, J=8.0 Hz, J=2.5 Hz, 1H), 7.76 (s, 1H), 7.73-7.68
(m, 2H), 3.13 (t, J = 6.5 Hz, 2H), 2.85 (m, 1H), 2.26 (d, J =
2.0 Hz, 3H), 1.87 (m, 1H), 0.92 (d, J = 7.0 Hz), 0.70 (m, 2H), 0.58
(m, 2H). LCMS: MHþ 370, retention time 2.86 min. Anal.
Calcd. for C21H24FN3O2: C, 68.27; H, 6.55; N, 11.37. Found:
C, 68.10; H, 6.52; N, 11.31.
6-Chloro-N-(2-methylpropyl)-3-pyridinecarboxamide (31). Com-
pound 31 was prepared in a similar manner to 30, using 2-methyl-
propylamine. 1H NMR (DMSO-d6) δ 8.83 (d, J = 2.0 Hz, 1H),
8.73 (bt, 1H), 8.23 (dd, J = 8.4 Hz, J = 2.4 Hz, 1H), 7.64 (d, J =
8.3 Hz, 1H), 3.09 (t, J = 6.0 Hz, 2H), 1.84 (m, 1H), 0.89 (d, J =
6.8 Hz, 6H).
6-{5-[(Cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}-
N-[(1R)-1,2-dimethylpropyl]-3-pyridinecarboxamide (15). 6-Chloro-
[(R)-(-)-3-methyl-2-butyl]nicotinamide (32, 100 mg, 0.44 mmol),
{5-[(cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}boronic
acid (26a, 100 mg, 0.42 mmol), tetrakis(triphenylphosphine)-
palladium (10 mg, 8.7 ꢀ 10-3 mmol), and aq sodium hydrogen
carbonate (4 mL) were mixed in propan-2-ol (8 mL) and heated at
90 °C under nitrogen for 18 h. The solvents were evaporated from the
cooled reaction under vacuum and the residue dissolved as far as
possible in ethyl acetate. The solution was applied to an SCX-2 ion
exchange cartridge (10 g) and the cartridge washed with ethyl acetate.
The product was eluted with 0.880 ammonia/methanol and the
solvents evaporated under vacuum. The residue was redissolved in
ethyl acetate and filtered through a silica cartridge (0.5 g). The filtrate
was reduced to dryness under vacuum and the residue triturated with
diethyl ether to give the desired compound as a white solid. 1H NMR
(DMSO-d6) δ9.10(d,J= 1.0 Hz, 1H), 8.57 (d, J= 4.0 Hz, 1H), 8.43
(d, J = 8.0 Hz, 1H), 8.32 (dd, J = 8.0 Hz, J = 2.0 Hz, 1H), 7.76
(s, 1H), 7.72-7.68 (m, 2H), 3.88 (m, 1H), 2.86 (m, 1H), 2.26 (d, J =
1.5 Hz, 3H), 1.79 (m, 1H), 1.14 (d, J = 7.0 Hz, 3H), 0.92 (d, J =
7.0 Hz, 6H), 0.70 (m, 2H), 0.57 (m, 2H). LCMS: MHþ 384, retention
time 2.93 min.
6-Chloro-[(R)-(-)-3-methyl-2-butyl]nicotinamide (32). 6-Chloro-
3-pyridinecarboxylic acid (3.0 g, 19 mmol) in thionyl chloride
(8.0 mL, 110 mmol) was heated at 90 °C for 2 h. The excess thionyl
chloride was removed under vacuum and the residue dissolved in
DCM (30 mL). A portion (5 mL) of this mixture was treated with
(R)-(-)-3-methyl-2-butylamine (0.50 mL), sodium carbonate
(1.0 g), and DCM (5 mL). The reaction was stirred at ambient
temperature overnight, filtered, and the filtrate reduced to dryness
under vacuum. The residue was further dried under vacuum to give
the desired product as a solid. 1H NMR (DMSO-d6) δ 8.82 (d, J =
2.5 Hz, 1H), 8.43 (d, J = 8.3 Hz, 1H), 8.24 (dd, J = 8.3 Hz, J =
2.5 Hz, 1H), 7.63 (d, J = 8.3 Hz, 1H), 3.83 (m, 1H), 1.76 (m, 1H),
1.10 (d, J = 6.8 Hz, 3H), 0.89 (dd, J = 6.8 Hz, J = 2.3 Hz, 6H).
LCMS: MHþ 227/229, retention time 2.60 min.
6-{5-[(Cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}-
N-(2,2-dimethylpropyl)-3-pyridinecarboxamide (2). 6-Chloro-
N-(2,2-dimethylpropyl)-3-pyridinecarboxamide (30, 560 mg,
2.3 mmol), N-cyclopropyl-5-fluoro-4-methyl-3-(4,4,5,5-tetra-
methyl-[1,3,2]dioxaborolan-2-yl)-benzamide (25a, 790 mg, 2.5
mmol), tetrakis(triphenylphosphine)palladium (30 mg, 0.025
mmol), and aq sodium hydrogen carbonate (1M, 3 mL) were
mixed in propan-2-ol (10 mL) and heated at 90 °C for 18 h. The
propan-2-ol was evaporated, the residue dissolved with ethyl
acetate, applied to an SCX-2 ion exchange cartridge (10 g), and
the cartridge washed with ethyl acetate and then methanol. The
product was eluted from the cartridge with 0.880 ammonia/ethyl
acetate/methanol and the solvents evaporated in vacuo. The
residue was triturated with diethyl ether to give the desired
product as a white solid (500 mg, 57%). The combined ethyl
acetate and methanol washings were reduced to dryness in
vacuo, the residue was dissolved in a mixture of acetone and
methanol and applied to an SCX-2 ion exchange cartridge (10 g),
and the cartridge was washed with methanol and eluted with
0.880 ammonia/methanol. The basic fraction was reduced to
dryness in vacuo to give a second batch of the desired product as
1
a white solid (200 mg, 22%). H NMR (DMSO-d6): δ 9.11 (d,
J = 1.5 Hz, 1H), 8.63 (t, J = 6.0 Hz, 1H), 8.56 (d, J = 4.0 Hz,
1H), 8.33 (dd, J = 8.0 Hz, J = 2.0 Hz, 1H), 7.76 (s, 1H),
7.73-7.68 (m, 2H), 3.16 (d, J = 6.5 Hz, 2H), 2.86 (m, 1H),
2.26 (d, J = 2.0 Hz, 3H), 0.93 (s, 9H), 0.70 (m, 2H), 0.58 (m, 2H).
LCMS: MHþ 384, retention time 3.01 min. Anal. Calcd. for
C22H26FN3O2: C, 68.9; H, 6.80; F, 4.95; N, 11.0. Found: C, 68.8;
H, 6.72; F, 5.03; N, 10.82.
6-Chloro-N-(2,2-dimethylpropyl)-3-pyridinecarboxamide (30).
6-Chloro-3-pyridinecarboxylic acid(500 mg, 3.1mmol)inthionyl
chloride (3.5 mL, 48 mmol) was heated at 100 °C for 1.5 h. The
excess thionyl chloride was evaporated under vacuum and the
solid residuedissolvedin DCM (20mL). Thismixture was treated
with 2,2-dimethylpropylamine (1.0 mL, 16 mmol) and sodium
carbonate (1.7 g, 16 mmol) and stirred for 15 h at ambient
temperature. The reaction was filteredandthe solvent evaporated
to give the title compound (840 mg, quant). 1H NMR (DMSO-d6)
δ 8.83 (d, J = 2.3 Hz, 1H), 8.61 (bt, 1H), 8.25 (dd, J = 8.3 Hz, J =
2.5 Hz, 1H), 7.64 (d, J = 8.3 Hz, 1H), 3.12 (d, J = 6.3 Hz, 2H),
0.90 (s, 9H).
{5-[(Cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}boronic
Acid (26a). A stirred solution of N-cyclopropyl-3-fluoro-5-iodo-4-
methylbenzamide (23a, 0.50 g, 1.6 mmol) in THF (20 mL) was
treated with sodium hydride (60% in mineral oil, 120 mg, 3.1 mmol).
The mixture was cooled to -75 °C and n-butyl lithium (1.6 M in
hexanes, 9.8 mL, 16 mmol) added slowly maintaining the tempera-
ture below -62 °C. Triisopropylborate (4.3 mL, 19 mmol) was
added dropwise over 10 min, maintaining the temperature below
-65 °C, and the mixture was stirred at -75 °C for 4 h and then
allowed to warm to ambient temperature overnight. The reaction
was recooled to -75 °C and quenched with water (20 mL). The
mixture was allowed to warm to ambient temperature and stirred for
30 min. The mixture was concentrated and the residue partitioned
between ethyl acetate and saturated ammonium chloride. The
aqueous phase was extracted with ethyl acetate (ꢀ2), the combined
organic phases dried (sodium sulfate), and the solvent evaporated
under vacuum. The residue was dissolved in ethyl acetate and the

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20 6267
solvent re-evaporated, and the residue was dissolved in DCM and
applied to a silica column. The column was eluted with an ethyl
acetate/cyclohexane gradient (50-100%) and then with methanol.
Appropriate fractions were reduced to dryness under vacuum to give
propan-2-ol (8 mL) and heated at 90 °C under nitrogen for 18 h.
The solvents were evaporated from the cooled reaction under
vacuum and the residue dissolved as far as possible in ethyl
acetate. The solution was applied to an SCX-2 ion exchange
cartridge (10 g) and the cartridge washed with ethyl acetate. The
product was eluted with 0.880 ammonia/methanol and the sol-
vents evaporated under vacuum. The residue was redissolved in
ethyl acetate and filtered through a silica cartridge (0.5 g). The
filtrate was reduced to dryness under vacuum and the residue
triturated with diethyl ether to give the desired compound as a
white solid. 1H NMR (DMSO-d6) δ 9.09 (d, J = 2.0 Hz, 1H), 8.57
(d, J = 4.0 Hz, 1H), 8.32-8.26 (m, 2H), 7.76 (s, 1H), 7.72-7.68
(m, 2H), 4.02 (m, 1H), 2.86 (m, 1H), 2.27 (d, J = 1.5 Hz, 3H), 1.12
(d, J=6.5 Hz, 3H), 0.93 (s, 9H), 0.70 (m, 2H), 0.58 (m, 2H).
LCMS: MHþ 398, retention time 3.05 min.
1
the desired compound (270 mg, 73%). H NMR (MeOD) δ 7.58
(s, 1H), 7.39 (d, J = 10.6 Hz, 1H), 2.82 (m, 2H), 2.31 (s, 3H), 0.79
(m, 2H), 0.63 (m, 2H). LCMS: MHþ 238, retention time 2.20 min.
6-{5-[(Cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}-
N-[(1S)-1,2-dimethylpropyl]-3-pyridinecarboxamide (16). 6-Chloro-
[(S)-(þ)-3-methyl-2-butyl]nicotinamide (33, 100 mg, 0.44 mmol),
{5-[(cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}boronic
acid (26a, 100 mg, 0.42 mmol), tetrakis(triphenylphosphine)-
palladium (10 mg, 8.7 ꢀ 10-3 mmol), and aq sodium hydrogen
carbonate (4.0 mL) were mixed in propan-2-ol (8 mL) and heated at
90 °C under nitrogen for 18 h. The solvents were evaporated from the
cooled reaction under vacuum and the residue dissolved as far as
possible in ethyl acetate. The solution was applied to an SCX-2 ion
exchange cartridge (10 g) and the cartridge washed with ethyl acetate.
The product was eluted with 0.880 ammonia/methanol and the
solvents evaporated under vacuum. The residue was redissolved in
ethyl acetate and filtered through a silica cartridge (0.5 g). The filtrate
was reduced to dryness under vacuum and the residue triturated with
diethyl ether to give the desired compound as a white solid. 1HNMR
(DMSO-d6) δ 9.10 (s, 1H), 8.57 (d, J = 3.5 Hz, 1H), 8.43 (d, J = 8.5
Hz, 1H), 8.32 (dd, J=8.0Hz,J= 2.5 Hz, 1H), 7.76 (s, 1H), 7.72-7.68
(m,2H),3.88(m,1H),2.86(m,1H),2.26(d,J=1.5Hz,3H),1.79(m,
1H), 1.14 (d, J = 6.5 Hz, 3H), 0.92 (d, J = 7.0 Hz, 6H), 0.70 (m, 2H),
0.58 (m, 2H). LCMS: MHþ 384, retention time 2.93 min.
6-Chloro-N-[(1S)-1,2,2-trimethylpropyl]-3-pyridinecarboxa-
mide (35). Compound 35 was prepared in a similar manner to 32,
1
using (S)-3,3-dimethyl-2-butylamine. H NMR (DMSO-d6) δ
8.80 (d, J = 2.5 Hz, 1H), 8.26-8.21 (m, 2H), 7.63 (d, J = 8.3 Hz,
1H), 3.96 (m, 1H), 1.08 (d, J = 7.0 Hz, 3H), 0.90 (s, 9H). LCMS:
MHþ 241/243, retention time 2.79 min.
Protein Crystallography Method. The crystal structure of
human unphosphorylated p38R with compound 11 was deter-
mined by closely following methodology previously described.10
More specificially in this case an apo-crystal was soaked (and
simultaneously cryoprotected) with compound 11 at 0.25 mM
for ∼2 days in soaking buffer containing 10% glycerol prior to
data collection at 100 K (using an Oxford Cryostream). The
X-ray diffraction data were collected on an in-house Rigaku-
MSC RuH2R rotating anode X-ray generator with a RAXIS
IVþþ image-plate detector. The data were processed with HKL
data processing package25 and CCP4 program suite26 and the
structure solved using the native p38 coordinates (PDB entry
1WFC) as the initial model in refinement by REFMAC.27 The
final R factor and R-free achieved for the complex were 18.1%
and 24.9%, respectively and the coordinates have been depos-
ited in the PDB as entry 3iph.
p38r Kinase Assay, Fluorescence Polarization Method. Fluor-
escence polarization assays used GST-tagged p38R (activated
using MKK6 and repurified) and an ATP-competitive rhodamine-
green labeled fluoroligand (2-(6-amino-3-imino-3H-xanthen-9-
yl)-5-{[({4-[4-(4-Cl-3-hydroxyphenyl)-5-(4-pyridinyl)-1H-imida-
zol-2yl]phenyl}methyl)amino]carbonyl}benzoic acid). These com-
ponents were dissolved in a buffer of final composition 62.5 mM
Hepes, pH 7.5, 1.25 mM CHAPS, 1 mM DTT, 12.5 mM MgCl2,
with final concentrations of 12 nM of p38R and 5 nM fluoroligand.
Then 30 μL of this mixture were added to wells containing 1 μL of
test compound (0.28 nM-16.6 μM) and incubated for 30-60 min
at room temperature. Fluorescence anisotropy was read in a
Molecular Devices Acquest (excitation 485 nm/emission 535 nm).
The Cheng-Prusoff equation (Cheng, Y.-C.; Prusoff, W. H.
Biochem. Pharmacol. 1973, 22, 3099-3108), Ki = IC50(1 þ
([S]/Km)), wasusedtocalculateKi from determined IC50 for activity
assays. To calculate Ki from determined IC50 for fluorescence
polarization assays a modification of the Cheng-Prusoff equation
was used (Cheng, H. C. Pharmacol. Res. 2005, 50, 21-40). Ki =
IC50(1 þ ((n[E])/Kd)) where n is the fraction of enzyme competent
to bind fluoroligand. Ki values for 381 compounds in the fluores-
cence polarization assay correlate with those from an activity assay
with r2 = 0.9.
6-Chloro-[(S)-(þ)-3-methyl-2-butyl]nicotinamide (33). Com-
pound 33 was prepared in a similar manner to 32, using (S)-(þ)-
3-methyl-2-butylamine. 1H NMR (DMSO-d6) δ 8.82 (d, J =
2.3 Hz, 1H), 8.42 (d, J = 8.5 Hz, 1H), 8.24 (dd, J = 8.3 Hz, J =
2.3 Hz, 1H), 7.64 (d, J = 8.3 Hz, 1H), 3.83 (m, 1H), 1.75 (m, 1H),
1.10 (d, J = 6.8 Hz, 3H), 0.89 (dd, J = 6.9 Hz, J = 2.4 Hz, 6H).
LCMS: MHþ 227/229, retention time 2.60 min.
6-{5-[(Cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}-
N-[(1R)-1,2,2-trimethylpropyl]-3-pyridinecarboxamide (17). 6-
Chloro-N-[(1R)-1,2,2-trimethylpropyl]-3-pyridinecarboxamide
(34, 100 mg, 0.42 mmol), {5-[(cyclopropylamino)carbonyl]-3-
fluoro-2-methylphenyl}boronic acid (26a, 100 mg, 0.42 mmol),
tetrakis(triphenylphosphine)palladium(10mg, 8.7ꢀ 10-3 mmol),
and aq sodium hydrogen carbonate (4.0 mL) were mixed in
propan-2-ol (8 mL) and heated at 90 °C under nitrogen for
18 h. The solvents were evaporated from the cooled reaction
under vacuum and the residue dissolved as far as possible in ethyl
acetate. The solution was applied to an SCX-2 ion exchange
cartridge (10 g) and the cartridge washed with ethyl acetate. The
product was eluted with 0.880 ammonia/methanol and the sol-
vents evaporated under vacuum. The residue was redissolved in
ethyl acetate and filtered through a silica cartridge (0.5 g). The
filtrate was reduced to dryness under vacuum and the residue
triturated with diethyl ether to give the desired compound as a
white solid. 1H NMR (DMSO-d6) δ 9.09 (d, J = 2.0 Hz, 1H), 8.57
(d, J = 4.0 Hz, 1H), 8.32-8.26 (m, 2H), 7.76 (s, 1H), 7.72-7.68
(m, 2H), 4.02 (m, 1H), 2.86 (m, 1H), 2.27 (s, 3H), 1.12 (d, J =
7.0 Hz, 3H), 0.93 (s, 9H), 0.70 (m, 2H), 0.57 (m, 2H). LCMS:
MHþ 398, retention time 3.05 min.
6-Chloro-N-[(1R)-1,2,2-trimethylpropyl]-3-pyridinecarboxamide
(34). Compound 34 was prepared in a similar manner to 32, using
1
(R)-3,3-dimethyl-2-butylamine. H NMR (DMSO-d6) δ 8.80 (d,
J = 2.5 Hz, 1H), 8.26-8.21 (m, 2H), 7.63 (d, J = 8.5 Hz, 1H), 3.96
(m, 1H), 1.08 (d, J = 6.8 Hz, 3H), 0.90 (s, 9H). LCMS: MHþ 241/
243, retention time 2.77 min.
6-{5-[(Cyclopropylamino)carbonyl]-3-fluoro-2-methylphenyl}-
N-[(1S)-1,2,2-trimethylpropyl]-3-pyridinecarboxamide (18). 6-
Chloro-N-[(1S)-1,2,2-trimethylpropyl]-3-pyridinecarboxamide
(35, 100 mg, 0.42 mmol), {5-[(cyclopropylamino)carbonyl]-3-
fluoro-2-methylphenyl}boronic acid (26a, 100 mg, 0.42 mmol),
tetrakis(triphenylphosphine)palladium(10mg, 8.7ꢀ 10-3 mmol),
and aq sodium hydrogen carbonate (1M, 4.0 mL) were mixed in
Cytokine Production in Human PBMC. PBMC were prepared
from heparinized human blood from normal volunteers by
centrifugation on hystopaque at 1000g for 30 min. The cells
were collected from the interface, washed by centrifugation
(1300g, 10 min) and resuspended in assay buffer (RPMI1640
containing 10% fetal calf serum, 1% L-glutamine, and 1%
penicillin/streptomycin) at 5 ꢀ 105 cells/mL. Then 100 μL cells
were added to microtiter wells containing 0.5 or 1.0 μL of
an appropriately diluted compound solution. After the assay
plates had been incubated for 1 h at 37 °C, 5% CO2, 25 μL of

6268 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20
Aston et al.
LPS (1 ng/mL final) was added and the samples incubated at
37 °C, 5% CO2 for a further 20 h. The supernatant was removed
and the concentrations of TNFR determined by enzyme linked
immunosorbant assay (ELISA) or using a multiplex bead
technology (Luminex) or antibody complex and electrochemi-
luminescence (Igen).
Cytokine Production in Human Whole Blood. Heparinized
blood drawn from normal volunteers was dispensed (100 μL)
into microtiter plate wells containing 0.5 or 1.0 μL of an
appropriately diluted compound solution. After 1 h incubation
at 37 °C, 5% CO2, 25 μL of LPS solution in RPMI 1640
(containing 1% L-glutamine and 1% penicillin/streptomycin)
was added (50 ng/mL final). The samples were incubated at
37 °C, 5% CO2 for 20 h, 100-150 μL physiological saline
(0.138% NaCl) was added, and diluted plasma was collected
using a Platemate or Biomek FX liquid handling robot after
centrifugation at 1300g for 10 min. Plasma TNFR content was
determined by enzyme linked immunosorbant assay (ELISA) or
using a multiplex bead technology (Luminex).
Supporting Information Available: NMR spectrum for com-
pound 2, pharmacokinetic protocols for rat cassette, rat dis-
crete, and cynomolgus monkey studies, rat PG-PS model
protocol, murine CIA protocol, poster giving further informa-
tion about modeling used in generation of virtual arrays. This
material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Taylor, P. C.; Williams, R. O.; Maini, R. N. Immunotherapy for
rheumatoid arthritis. Curr. Opin. Immunol. 2001, 13 (5), 611–616.
(2) 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.; Strickler, J. E.; McLaughlin, M. M.; Sie-
mens, I. R.; Fisher, S. M.; Livi, G. P.; White, J. R.; Adams, J. L.;
Young, P. R. A protein kinase involved in the regulation of
inflammatory cytokine biosynthesis. Nature 1994, 372, 739–746.
(3) Herlaar, E.; Brown, Z. p38 MAPK signalling cascades in inflam-
matory disease. Mol. Med. Today 1999, 5, 439–447.
(4) Peifer, C.; Wagner, G.; Laufer, S. New approaches to the treatment
of inflammatory disorders small molecules inhibitors of p38 MAP
kinase (Review). Curr. Top. Med. Chem. 2006, 6, 113–149.
(5) Goldstein, D. M.; Gabriel, T. Pathway to the clinic: Inhibition of
p38 MAP kinase. A review of ten chemotypes selected for devel-
opment (Review). Curr. Top. Med. Chem. 2005, 5, 1017–1029.
(6) McClure, K. F.; Letavic, M. A.; Kalgutkar, A. S.; Gabel, C. A.;
Audoly, L.; Barberia, J. T.; Braganza, J. F.; Carter, D.; Carty, T. J.;
Cortina, S. R.; Dombroski, M. A.; Donahue, K. M.; Elliott, N. C.;
Gibbons, C. P.; Jordan, C. K.; Kuperman, A. V.; Labasi, J. M.;
LaLiberte, R. E.; McCoy, J. M.; Naiman, B. M.; Nelson, K. L.;
Nguyen, H. T.; Peese, K. M.; Sweeny, F. J.; Taylor, T. J.; Trebino, C.
E.; Abramov, Y. A.; Laird, E. R.; Volberg, W. A.; Zhou, J.; Bach, J.;
Lombardo, F. Structure-activity relationships of triazolopyridine
oxazole p38 inhibitors: identification of candidates for clinical
development. Bioorg. Med. Chem. Lett. 2006, 16 (16), 4339–4344.
(7) Natarajan, S. R.; Wisnoski, D. D.; Thompson, J. E.; O’Neill, E. A.;
O’Keefe, S. J. p38 MAP kinase inhibitors. Part 3: SAR on 3,4-
dihydropyrimido[4,5-d]pyrimidones and 3,4-dihydropyrido[4,3-
d]pyrimidin-2-ones. Bioorg. Med. Chem. Lett. 2006, 16 (16),
4400–4404.
(8) Dominguez, C.; Powers, D. A.; Tamayo, N. p38 MAP kinase
inhibitors: Many are made, but few are chosen. Curr. Opin. Drug
Dis. Dev. 2005, 8 (4), 421–430.
(9) Lee, M. R.; Dominguez, C. MAP kinase p38 inhibitors: clinical
results and an intimate look at their interactions with p38r protein.
Curr. Med. Chem. 2005, 12, 2979–2994.
(10) Angell, R. M.; Bamborough, P.; Cleasby, A.; Cockerill, S. G.;
Jones, K. L.; Mooney, C. J.; Somers, D. O.; Walker, A. L. Biphenyl
amide p38 kinase inhibitors: discovery and binding mode. Bioorg.
Med. Chem. Lett. 2008, 18 (1), 318–323.
(11) Angell, R. M.; Angell, T. D.; Bamborough, P.; Brown, D.; Brown,
M.; Buckton, J. B.; Cockerill, S. G.; Edwards, C. D.; Jones, K. L.;
Longstaff, T.; Smee, P. A.; Smith, K. J.; Somers, D. O.; Walker, A.
L.; Willson, M. Biphenyl amide p38 kinase inhibitors. 2: Optimiza-
tion and SAR. Bioorg. Med. Chem. Lett. 2008, 18 (1), 324–328.
(12) Angell, R. M.; Aston, N. M.; Bamborough, P.; Buckton, J. B.;
Cockerill, S. G.; de Boeck, 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 (15), 4428–4432.
(13) Angell, R. M.; Angell, T. D.; Bamborough, P.; Bamford, M. J.;
Chung, C.; Cockerill, S. G.; Flack, S. S.; Jones, K. L.; Laine, D. I.;
Longstaff, T.; Ludbrook, S.; Pearson, R.; Smith, K. J.; Smee, P. A.;
Somers, D. O.; Walker, A. L. Biphenyl amide 38 kinase inhibitors.
4: DFG-in and DFG-out binding modes. Bioorg. Med. Chem. Lett.
2008, 18 (15), 4433–4437.
TNFr ELISA. Human TNFR was assayed using a commer-
cial assay kit (AMS Biotechnology 211-90-164-40) according to
the manufacturer’s instructions but with TNFR calibration
curves prepared using Pharmingen TNFR (cat. no. 555212).
LUMINEX Assay. First, 45 μL of luminex assay buffer (PBS/
1%BSA/0.025% Triton X-100) was added to 96-well microtiter
Millipore filter plates (MABVN1210) followed by 5 μL of HWB
or PBMC supernatant. Then 50 μL of Luminex microspheres
coated in a primary antibody against human TNFR (1 ꢀ 105
beads/mL) were added and the plates were shaken and then
incubated at 4 °C overnight in the dark. Standard curves for
TNFR were present on each plate (0-2000 pg/mL for TNFR).
Plates were then washed under low vacuum filtration and 50 μL/
well of diluted biotinylated secondary antibodies added. The
plates were shaken for 1 h at room temperature, washed, and
then 50 μL/well of streptavidin phycoerythrin added (Zymed
15-4301) at 0.25 μg/mL. The plates were shaken for 30 min in the
dark, and after a final wash, the beads were resuspended in
120 μL of Sheath fluid (BioRad cat. no. 171-000055) and the
plates read on the BioRad multiplex reader.
Data Analysis. TNFR concentrations were determined from
standard curves, and IC50s were generated from percent inhibi-
tion values using nonlinear least-squares curve fitting, in general
with a four-parameter logistic model using Activity Base.
CYP450 Assay. The inhibition of recombinant CYP450
mediated O-dealkylase metabolism of pro-fluorescent probe sub-
strates was assessed in a fluorimetric assay with the following
recombinant human cytochrome P4500s coexpressed in Escherichia
coli with human NADPH reductase (Bactosomes, CYPEX UK),
CYP2C9 with the substrate 7-methoxy-4-trifluoromethylcoumar-
in-3-acetic acid (FCA), CYP1A2 with the substrate ethoxyresor-
ufin (ER), CYP2D6 with the substrate 4-methylaminomethyl-
7-methoxycoumarin (MMMC), CYP2C19 with the substrate
3-butyryl-7-methoxycoumarin (BMC), and CYP3A4 with the
substrates 7-benzyloxyquinolone (7BQ) and diethoxyfluorescein
(DEF). Compounds were incubated at a starting concentration of
5 mM in <2% MeOH. The concentration range for the IC50
determination assay was 0.1-100 uM (spread over nine points)
and included a 0.2 mM NADPH regeneration system.
(14) Schrier, D. C.; Schimmer, R. C.; Flory, C. M.; Tung, D.K-L.;
Ward, P. A. Role of chemokines and cytokines in a reactivation
model of arthritis in rats induced by injection with streptococcal
cell walls. J. Leukocyte Biol. 1998, 63, 359–363.
(15) Courtenay, J. S.; Dallman, M. J.; Dayan, A. B.; Martin, A.;
Mosedale, B. Immunisation against heterologous type II col-
lagen-induced arthritis in mice. Nature 1980, 2843, 666–668.
(16) Joosten, L. A. B.; Helsen, M. M. A.; van de Loo, F. A. J.; van den
Berg, W. B. Anticytokine treatment of established collagen-in-
duced arthritis in DBA/1 mice. Arthritis Rheum. 1996, 39, 797–809.
(17) Joosten, L. A. B.; Helsen, M. M. A.; Saxne, T.; van de Loo, F. A. J.;
Heinegard, D.; van den Berg, W. B. IL-1rβ blockade prevents
cartilage and bone destruction in murine type II collagen induced
Acknowledgment. We thank Antony Cooper, Suzanne J.
deBoeck, Derek Evans, Kathryn J. Smith, Stephen Swanson,
and Aurelien Putey for synthetic and medicinal chemistry
contributions, Antony D. R. Angell, Yemisi E. Solanke, and
Richard A. Williamson for generation of the PBMC and
HWB data, Chris Mooney for generation of p38R protein,
Margarete Neu for her contribution to the crystallography
studies, Klara Valko for generation of CHI logD data, and
Chris Luscombe for the statistical modeling. This work was
funded by GlaxoSmithKline.

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 20 6269
arthritis whereas TNF-r blockade only ameliorates joint inflam-
mation. J. Immunol. 1999, 163, 5049–5055.
(23) Okada, S.; Kayakiri, N.; Sawaday, Y.; Tanaka, H.; Hashimoto, M.
Indole derivatives. World Patent WO93/03012, 1993.
(18) Williams, R. O.; Feldmann, M.; Maini, R. N. Anti-tumor necrosis
factor ameliorates joint disease in murine collagen-induced arthri-
tis. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 9784–9788.
(19) Valko, K.; Bevan, C.; Reynolds, D. Chromatographic Hydropho-
bicity Index by Fast-Gradient RP-HPLC: A High-Throughput
Alternative to Log P/log D. Anal. Chem. 1997, 69 (11), 2022–2029.
(20) HSA binding constants were obtained by high performance affi-
nity chromatography on a column of immobilized human serum
albumin.
(21) Solubilities were measured either from dissolution of solid into
water buffered to pH7.4 or by precipitation from 10 mM DMSO
solution.
(22) Edwards C. D. Drug Metabolism Discussion Group, Sept 2004.
Poster included in Supporting Information.
(24) Schlosser, M.; Geneste, H. The super base approach to flurbipro-
fen: an exercise in optionally site-selective metalation. Chem.;Eur.
J. 1998, 4 (10), 1969–1973.
(25) Otwinowski. Z.; Minor, W. Processing of X-ray Diffraction
Data Collected in Oscillation Mode. Macromolecular Crystallo-
graphy, Part A; Methods in Enzymology, Vol. 276; Carter, C.W., Jr.,
Sweet, R. M., Eds.; Academic Press: New York, 1997; pp 307-
326.
(26) Collaborative Computational Project Number 4. The CCP4 Suite:
Programs for Protein Crystallography. Acta Crystallogr., Sect. D:
Biol. Crystallogr. 1994, 50 (5), 760–763.
(27) 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 (5), 240–255.