Synthesis and pharmacological characterization of a potent, orally active p38 kinase inhibitor.

Article abstract of DOI:10.1016/S0960-894X(02)00238-X

Inhibitors of the MAP kinase p38 provide a novel approach for the treatment of osteoporosis, inflammatory disorders, and cancer. We have identified N-(3-tert-butyl-1-methyl-5-pyrazolyl)-N'-(4-(4-pyridinylmethyl)phenyl)urea as a potent and selective p38 ki

Full text of DOI:10.1016/S0960-894X(02)00238-X

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1559–1562
Synthesis and Pharmacological Characterization of a Potent,
Orally Active p38 Kinase Inhibitor
Jacques Dumas,a,* Holia Hatoum-Mokdad,a Robert N. Sibley,a Roger A. Smith,a
William J. Scott,a Uday Khire,a Wendy Lee,a Jill Wood,a Donald Wolanin,a
Jeffrey Cooley,a Donald Bankston,a Aniko M. Redman,a Robert Schoenleber,a
Yolanda Caringal,a David Gunn,a Romulo Romero,a Martin Osterhout,a
Holger Paulsen,a Timothy J. Housley,b Scott M. Wilhelm,b John Pirro,c
Du-Shieng Chien,c Gerald E. Ranges,b Alka Shrikhande,b Andrew Muzsi,b
Elizabeth Bortolon,b Jean Wakefield,b Cynthia Gianpaolo Ostravage,b Ajay Bhargavab
and Thuy Chaub
aDepartment of Chemistry Research, Bayer Research Center, 400 Morgan Lane, West Haven, CT 06516, USA
bDepartment of Cancer and Osteoporosis Research, Bayer Research Center, 400 Morgan Lane, West Haven, CT 06516, USA
cInstitute of Preclinical Drug Development, Bayer Research Center, 400 Morgan Lane, West Haven, CT 06516, USA
Received 26 March 2002; accepted 16 April 2002
Abstract—Inhibitors of the MAP kinase p38 provide a novel approach for the treatment of osteoporosis, inflammatory disorders,
and cancer. We have identified N-(3-tert-butyl-1-methyl-5-pyrazolyl)-N0-(4-(4-pyridinylmethyl)phenyl)urea as a potent and selective
p38 kinase inhibitor in biochemical and cellular assays. This compound is orally active in two acute models of cytokine release
(TNF-induced IL-6 and LPS-induced TNF) and a chronic model of arthritis (20-day murine collagen-induced arthritis). # 2002
Elsevier Science Ltd. All rights reserved.
The MAP kinase p38 is involved in IL-1b and TNFa
signaling pathways,1 and provides a novel approach to
the treatment of osteoporosis and inflammatory dis-
orders. For example, SB 203580 (1, Fig. 1), shows
potent activity in models of endotoxin shock and bone
resorption.2 In addition, p38 kinase has been linked to
angiogenesis through cellular VEGF production.3
Screening of a Bayer combinatorial chemistry library
afforded a novel class of p38 inhibitors such as urea 2.
further improvements of both potency and drug-like
properties.6
Synthetic routes to ureas similar to 2 and 3, as well as
the SAR of the pyrazole unit of urea 2 have already
been reported by our group.4,7 The focus of this study is
the bisaryl ether side chain of 3 (Tables 1 and 2). While the
introduction of nonpolar substituents at the 4-position of
Early optimization of urea 2 (Fig. 1) into phenylpyr-
azolyl ureas has already been reported in this journal,
and provided in vivo proof-of-principle for this com-
pound class.4 More recently, X-ray crystallographic
data suggested that diarylureas are binding to an allos-
teric domain of p38 kinase.5 The present report focuses
on optimization of phenoxy-phenyl urea 3, resulting in
*Corresponding author. Tel.: +1-203-812-5204; fax: +1-203-812-
3655; e-mail: jacques.dumas.b@bayer.com
Figure 1. P38 kinase inhibitors.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00238-X

1560
J. Dumas et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1559–1562
the phenoxy group (entries 4–6) does not seem to
impact p38 kinase activity, addition of hydrogen bond
donors such as amides (7–9), carbamates (11–12), or
phenols (13) improves potency. Pyridines 22 and 23
(Table 2) also offer good replacements for the phenyl
group of 3 and display potent activity in both the p38
kinase assay and the functional assay (IL-1–TNFa-
induced IL-6 production in SW1353 cells).8 The oxygen
linker atom of urea 3 can easily be replaced by a sulfur
atom, a methylene, or a thiomethylene group. The
observation that the length of the linker is non-critical
for activity, and that hydrophilic substituents are pre-
ferred on this side of the molecule suggests that the
extremity of the biaryl side chain could be extending
into solvent. Our first acute in vivo model measures
TNFa induced IL-6 production in mice, and provides
oral efficacy data with a rapid turnaround time.9 Car-
bamate 11 shows a promising in vitro profile, but does
not demonstrate significant activity in our in vivo
screening model (Table 3). In contrast, urea 23 and its
methylene analogue 22 show significant effects in this
model, at doses ranging from 25 to 50 mg/kg po. The
latter analogue was selected for chronic pharmacology
studies.
The in vitro kinase selectivity of urea 22 was evaluated
against a panel of receptor kinases and cytosolic kina-
ses. Of these, only the isoform p38 b1 is inhibited at
appreciable levels.10 This finding appears consistent
with the hypothesis developed by Regan et al.5 in which
pyrazolyl ureas bind to an allosteric stire specific to p38,
outside of the ATP pocket. Plasma exposure following
single dose oral administration in mice was then mea-
sured, and is depicted in Figure 2. Micromolar levels of
22 are maintained for 4 h after a 10 mg/kg single oral
administration (0.5% Tween 80/water vehicle). The
ability of urea 22 to block LPS-induced TNFa synth-
esis11 was also measured, and is summarized in Table 4.
Potent inhibition of TNFa production is observed at
20 mg/kg po single dose. In comparison, SB 203580 1 is
also efficacious in this model (30 mg/kg po).
Table 1. Substitution of the bisaryl ether
The activity of urea 22 was then assessed in a 20-day,
murine model of arthritis (Fig. 3, Table 5).12 In this
model, mice are immunized with bovine type II col-
lagen, then treated with test compounds or vehicle. The
onset of arthritis is then measured using a variety of
parameters, including a clinical score, and three histo-
pathology readouts. Both compounds 1 and 22 showed
oral efficacy in this model at 50 and 30 mg/kg po daily
dose, respectively.
Compd
X
R
p38 a2
(IC50, nM)
SW1353
(IC50, nM)
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
O
O
O
O
O
O
O
O
O
O
S
S
S
CH2
CH2
CH2
CH2
CH2
CH2
H
CH3
tert-Bu
OCH2Ph
270
160
183
110
235
45
69
310
33
54
35
57
69
37
87
290
70
Table 3. Activity of selected analogues in TNFa-induced IL-6 short-
term in vivo model
NHCOCH3
NHCOCH2CH(CH3)2
NHCOCH2CH3
NH2
313
1060
Compd Aqueous solubility (mg/L)
Dose
(mg/kg po) Inhib. Value
%
p
NHCO2Et
NHCO2-iPr
OH
87
162
1100
235
>2500
170
57
pH 7.5
pH 2.0
1
100
50
25
100
50
50
50
80
84
60
0.001
0.003
0.01
OnPr
OnBu
NHCOCH2CH(CH3)2
NHCO2-tert-Bu
NH2
NHCOCH3
NHCOCH2CH3
NHCO(CH2)2CO2H
22
22
23
23
11
25
25
25
50
2000
1658
655
90 <0.001
77 <0.001
16
64 <0.001
22
ꢀ15
95
322
77
5155
NS
38
200
1750
>2500
25
10
0.025
NS
Table 2. Substitution of the bisaryl ether: linker atom
Compd
X
p38 a2
(IC50, nM)
SW1353
(IC50, nM)
22
23
24
25
26
27
28
CH2
S
SCH2
CH2S
NH
42
13
140
44
420
430
260
323
95
584
CH2CH2
OCH2
Figure 2. Plasma exposure of urea 22, following a single-dose oral
administration in mice.

J. Dumas et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1559–1562
1561
Table 4. Activity of compounds 1 and 22 in LPS-induced TNFa
production in mice
Table 5. Activity of SB 203580 (1) and urea 22 in 20-day murine
collagen-induced arthritis model: clinical scores and histopathology
Compd
Dose
(mg/kg po)
Plasma
TNFa
%
Inhib.
p
Value
Compd
Dose
(mg/kg)
Clinical score
Histopathology
(ng/mL,ꢂSEM)
No. paws affected % regression
IC
IO
Vehicle
1
1
22
22
22
37.5ꢂ2.9
38.0ꢂ5.6
13.0ꢂ2.8
46.7ꢂ5.3
19.8ꢂ2.3
5.1ꢂ0.7
Vehicle
1
22
22
22
2.3ꢂ0.4
1.0ꢂ0.4a
1.4ꢂ0.5
1.9ꢂ0.5
0.7ꢂ0.3a
19.1ꢂ8.0
1.13
0.63
10
30
10
20
50
ꢀ1.4
65.4
NS
<0.0001
NS
<0.0001
<0.0001
50
3
10
30
65.0ꢂ13.5a
43.5ꢂ15.8
25.0ꢂ13.4
61.7ꢂ14.5a
0.63a 0.25a
0.80
0.78
0.38a 0.20a
0.33
0.30
ꢀ24.5
47.1
86.5
IC, inflammatory cells; IO, interstitial oedema.
aSignificant from vehicle (p<0.05).
82, 591. Jackson, J. R.; Bolognese, B.; Hillegass, L.; Kassis, S.;
Adams, J.; Griswold, D. E.; Winkler, J. D. J. Pharmacol. Exp.
Ther. 1998, 284, 687.
4. Dumas, J.; Sibley, R.; Riedl, B.; Monahan, M.-K.; Lee, W.;
Lowinger, T. B.; Redman, A. M.; Johnson, J. S.; Kingery-
Wood, J.; Scott, W. J.; Smith, R. A.; Bobko, M.; Schoenleber,
R.; Ranges, G. E.; Housley, T. J.; Bhargava, A.; Wilhelm,
S. M.; Shrikhande, A. Bioorg. Med. Chem. Lett. 2000, 10,
2047. Ranges, G. E.; Bortolon, E.; Chau, T.; Dixon, B. R.;
Bhargava, A.; Dumas, J.; Gianpaolo-Ostravage, C.; Hatoum-
Mokdad, H.; Housley, T. J.; Shrikhande, A.; Scott, W. J.;
Sibley, R.; Wakefield, J.; Wilhelm, S. M. 220th ACS National
Meeting, Washington, DC, Aug 20–24, 2000; MEDI-149.
5. Pargellis, C.; Tong, L.; Churchill, L.; Cirillo, P. F.; Gil-
more, T.; Graham, A. G.; Grob, P. M.; Hickey, E. R.; Moss,
N.; Pav, S.; Regan, J. Nat. Struct. Biol. 2002, 9, 268.
Figure 3. Activity of SB 203580 (1) and urea 22 in 20-day murine
collagen-induced arthritis model: severity scores.
6. Dumas, J.; Hatoum-Mokdad, H.; Sibley, R.; Smith, R. A.;
Scott, W. J.; Khire, U.; Lee, W.; Wood, J.; Wolanin, D.;
Cooley, J.; Bankston, D.; Redman, A.; Schoenleber, R.;
Caringal, Y.; Housley, T. J.; Wilhelm, S. M.; Pirro, J.; Chien,
D. S.; Ranges, G. E.; Shrikhande, A.; Muzsi, A.; Bortolon, E.;
Wakefield, J.; Gianpaolo-Ostravage, C.; Chau, T. 222nd ACS
National Meeting, Chicago, IL, Aug 26–30, 2001; MEDI-256.
7. Redman, A. M.; Johnson, J. S.; Dally, R.; Swartz, S.; Wild,
H.; Paulsen, H.; Caringal, Y.; Gunn, D.; Renick, J.; Osterh-
out, M.; Kingery-Wood, J.; Smith, R. A.; Lee, W.; Dumas, J.;
Wilhelm, S. M.; Housley, T. J.; Bhargava, A.; Ranges, G. E.;
Shrikhande, A.; Young, D.; Bombara, M.; Scott, W. J. Bioorg.
Med. Chem. Lett. 2001, 11, 9.
In conclusion, optimization of urea 3 through a com-
bined combinatorial and medicinal chemistry effort
resulted in improvements of both potency and drug-like
properties. As a result, we identified N-(3-tert-butyl-1-
methyl-5-pyrazolyl)-N0-(4-(4-pyridinyl-methyl)phenyl)urea
2213 as a potent and selective inhibitor of the MAP
kinase p38. This analogue is orally available in mice,
and shows significant in vivo activity in two acute
models of cytokine release (TNFa-induced IL-6 and
LPS-induced TNFa) and a chronic model of arthritis
(20-day murine collagen induced arthritis).
8. P38 kinase in vitro assay: purified and His-tagged p38 a2
(expressed in Escherichia coli) was activated in vitro by MMK-
6 to a high specific activity. Using a microtiter format, all
reactions were conducted in 100 mL volumes with reagents
diluted to yield 0.05 mg/well of activated p38 a2 and 10 mg/well
of myelin basic protein in assay buffer (25 mM HEPES 7.4,
20 mM MgCl2, 150 mM NaCl). Test compounds (5 mL of a
10% DMSO solution in water) were prepared and diluted into
the assay to cover a final concentration range from 5 nM to
2.5 mM. The kinase assay was initiated by addition of 25 mL of
an ATP cocktail to give a final concentration of 10 mM cold
ATP and 0.2 mCi [g-33P] ATP per well (200–400 dpm/pmol of
ATP). The plate was incubated at 32 ꢁC for 35 min, and the
reaction quenched with 7 mL of a 1 N aq HCl solution. The
samples were harvested onto a P30 Filtermat (Wallac, Inc.)
using a TomTec 1295 Harvester (Wallac, Inc.), and counted in
a LKB 1205 Betaplate Liquid Scintillation Counter (Wallac,
Inc.). Negative controls included substrate plus ATP alone.
SW1353 cellular assay: SW1353 cells (human chondro-sar-
coma, ATCC, Bethesda, MD) are seeded (1000 cells/100 mL
DMEM 10% FCS/well) into 96-well plates and incubated
overnight. After medium replacement, cells are exposed to test
compounds for 1 h at 37 ꢁC, at which time human IL-1 (1 ng/
mL, Endogen, Woburn, WA) and recombinant human TNFa
(10 ng/mL) are added. Cultures are incubated for 48 h at
Acknowledgements
We would like to thank Christopher Browe, Tim Sarr,
and Robert Dreyer (cloning and purification of p38 a2
kinase and MKK-6), and Teresa Lopes (solubility mea-
surements) for their support. We are also grateful to
Charles Maniglia, David Brittelli, Timothy Lowinger,
Bernd Riedl, and Robert Carlson for helpful discussions.
References and Notes
1. Gallagher, T. F.; Fier-Thompson, S. M.; Garigipati, R. S.;
Sorenson, M. E.; Smietana, J. M.; Lee, D.; Bender, P. E.; Lee,
J. C.; Laydon, J. T.; Chabot-Fletcher, M. C.; Breton, J. J.;
Adams, J. L. Bioorg. Med. Chem. Lett. 1995, 5, 1171.
2. Badger, A. M.; Bradbeer, J. N.; Votta, B.; Lee, J. C.;
Adams, J. L.; Griswold, D. E. J. Pharm. Exper. Ther. 1996,
279, 1453.
3. Yamamoto, T.; Kozawa, O.; Tanabe, K.; Akamatsu, S.;
Matsuno, H.; Dohi, S.; Uematsu, T. J. Cell. Biochem. 2001,

1562
J. Dumas et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1559–1562
37 ꢁC, then supernatant IL-6 values are determined by ELISA
(Endogen, Woburn, WA). For both assays, at least two inde-
pendent IC50 determinations were performed on each compound,
and the mean value is reported. SB203580 (1) was used as a
standard (p38 IC50=20 nM, SW1353 IC50=50 nM). In all cases,
standard deviations were less than 50% of the mean IC50 value.
9. In vivo IL-6 induction by TNFꢀ: BALB/c female mice are
dosed po with test compounds and given an ip injection of
rHuTNF after 1 h. Two hours later, animals are bled by car-
diac puncture. Plasma IL-6 levels are determined by ELISA,
SB 203580 (1) is used as a positive control.
13. Large scale preparation of urea 22: Step 1: Synthesis of 5-
amino-3-tert-butyl-1-methylpyrazole. A solution of pivaloyl-
acetonitrile (320 g, 2.56 mol) in ethanol (3000 mL) was treated
with methylhydrazine (180 mL, 155.88 g, 3.38 mol), and the
contents were heated to 70 ꢁC for 18 h. The solution was then
cooled to room temperature and concentrated to a white solid,
which was triturated with hexane (2ꢃ2000 mL) to provide
5-amino-3-tert-butyl-1-methylpyrazole (373 g, 2.43 mol, 95%)
1
as colorless needles after vacuum drying. H NMR (CDCl3) d
1.26 (s, 9H, tert-butyl), 3.45 (br s, 2H, –NH2), 3.63 (s, 3H,
–NCH3), 5.42 (s, 1H, –CH). Anal. calcd for C8H15N3: C,
62.71; H, 9.87; N, 27.42. Found: C, 62.80; H, 9.66; N, 27.35.
MS/EI: m/z 153 (M+); 138 (MꢀCH3).
10. In vitro kinase selectivity of urea 22: this data was obtained
in house and at MDS Panlabs (Bothell, WA):
Step 2: Synthesis of 4-(4-pyridinylmethyl)benzenamine. A
mixture of 4-(4-nitrophenylmethyl)pyridine (200.0 g, 933.62
mmol) and 10% Pd/C (34 g, Degussa type) in 1000 mL of
ethanol/ethyl acetate (4:1) was shaken in a Parr vessel under
20–30 psi hydrogen pressure until reduction was complete by
TLC detection. The mixture was filtered over Celite, and the
filtrate was concentrated. The solids were triturated with hot
20% ethanol/hexane (1000 mL) to provide 4-(4-aminophenyl
methyl)pyridine (165.0 g, 895.57 mmol, 96%) as a light-brown
solid after vacuum drying. 1H NMR (DMSO-d6) d 3.74 (s, 2H,
–CH2–), 4.92 (br s, 2H, –NH2), 6.49, 6.87 (AA0BB0, J=8.4 Hz,
4H, phenyl), 7.17, 8.41 (AA0BB0, J=5.9 Hz, 4H, phenyl).
Anal. calcd for C12H12N2: C, 78.23; H, 6.57; N, 15.20. Found:
C, 77.94; H, 6.41; N, 15.15. MS/EI: m/z 184 (M+). Mp 156–
158 ꢁC.
p38 a2
p38 b1
HER-2
JNK-1
ERK-1
abl
p59lck
p59fyn
PKA
IC50=42 nM
IC50=110 nM
IC50=4.2 mM
2% inhibition at 0.5 mM
8% inhibition at 5 mM
3% inhibition at 5 mM
8% inhibition at 10 mM
20% inhibition at 10 mM
7% inhibition at 10 mM
0% inhibition at 10 mM
5% inhibition at 10 mM
21% inhibition at 10 mM
17% inhibition at 10 mM
PKCa
PKCb
PKCg
EGFR
11. In vivo TNFꢀ induction by LPS: Female BALB/c mice are
dosed po with test compounds (0.5% Tween 80/water vehicle).
After 1 h, animals are treated with 100 mg LPS (ip injection).
After 90 min, the animals are bled by cardiac puncture. TNFa
levels in plasma are determined by ELISA, SB 203580 (1) is
used as a positive control.
Step 3: synthesis of urea 22. A suspension of 5-amino-3-tert-
butyl-1-methylpyrazole (300.0 g, 1.96 mol) in methylene chlo-
ride (3000 mL) was treated with CDI (334.0 g, 2.06 mol, 1.05
equiv) in one portion, and the mixture was stirred at room
temperature under argon. After 15 min, dissolution occurred,
and the temperature of the solution slowly rose to 30 ꢁC. After
60 min, the reaction flask was charged with 4-(4-aminophe-
nylmethyl) pyridine (361.10 g, 1.96 mol) in one portion, and
the internal temperature slowly rose to a gentle reflux. When
the internal temperature cooled to 36 ꢁC, the contents were
stirred for 18 h at that temperature under an atmosphere of
argon. The reaction mixture was then cooled to room tem-
perature, washed with brine (3ꢃ2000 mL) and the organic
phase was dried (sodium sulfate) and concentrated to a tan
solid foam. The crude product was heated to reflux in ethyl
acetate (4000 mL) for 20 min, cooled to 30 ꢁC and treated with
diethyl ether (4000 mL). Suction filtration provided N-(3-tert-
butyl-1-methyl-5-pyrazolyl)-N0-(4-(4-pyridinylmethyl)phenyl)urea
12. Murine collagen induced arthritis in vivo model: this model
was performed at Chrysalis International, Inc. On day 0,
groups of 12 mice are immunized at the base of the tail with
100 mg of bovine type II collagen, which is emulsified with
complete Freund’s adjuvant. On day 7, a second booster dose
of collagen is administered the same way. On day 14, the mice
are treated with 100 mg LPS (sc). On day 21, at the onset of
arthritis, the mice are dosed po once daily with test com-
pounds or vehicle (0.5% Tween 80 in water) for 20 days. Paw
oedema is measured by a constant caliper, and arthritic mani-
festations are clinically scored during the 20-day observation
period. The severity score is the total clinical score (rated from
0 to 4 for each paw) divided by the total number of mice in the
group. Column ‘No. paws affected’ reflects the total number of
affected paws divided by the number of mice in the group. At
the end of the study, the paws are removed and examined
histologically (number of inflammatory cells, interstitial
oedema). Positive control dexamethasone (3 mg/kg po) totally
prevents the onset of arthritis in this model.
1
22 (601.0 g, 1.65 mol, 84%) as a pale-yellow solid. H NMR
(DMSO-d6) d 1.19 (s, 9H, t-butyl), 3.57 (s, 3H, –NCH3), 3.88
(s, 2H, –CH2–), 6.02 (s, 1H, –CH), 7.14, 7.38 (AA0BB0, 8.5 Hz,
4H, Ph), 7.21, 8.44 (AA0BB0, 5.9 Hz, 4H, phenyl), 8.43 (s, 1H,
–NH), 8.82 (s, 1H, –NH). MS/FAB: m/z 364 (M+1). Anal.
calcd for C21H25N5O: C, 69.40; H, 6.93; N, 19.27. Found: C,
69.04; H, 6.93; N, 19.34. Mp 161–162 ꢁC.