3-Amino-pyrazolo[3,4-d]pyrimidines as p38α kinase inhibitors: Design and development to a highly selective lead

Article abstract of DOI:10.1016/j.bmcl.2011.03.098

Learnings from previous Roche p38-selective inhibitors were applied to a new fragment hit, which was optimized to a potent, exquisitely selective preclinical lead with a good pharmacokinetic profile.

Full text of DOI:10.1016/j.bmcl.2011.03.098

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3452–3456
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
3-Amino-pyrazolo[3,4-d]pyrimidines as p38
and development to a highly selective lead
a kinase inhibitors: Design
Michael Soth a, , Sarah Abbot a, Allassan Abubakari a, Nidhi Arora a, Humberto Arzeno a, Roland Billedeau a,
Nolan Dewdney a, Kieran Durkin a, Sandra Frauchiger a, Manjiri Ghate a, David M. Goldstein a, Ronald J. Hill a,
Andreas Kuglstatter a, Fujun Li a, Brad Loe a, Kristen McCaleb a, Joel McIntosh a, Eva Papp a, Jaehyeon Park a,
Martin Stahl b, Man-Ling Sung a, Rebecca Suttman a, David C. Swinney a, Paul Weller a, Brian Wong a,
Hasim Zecic a, Tobias Gabriel a
⇑
a Roche Palo Alto, 3431 Hillview Avenue, Palo Alto, CA 94304, USA
b Discovery Chemistry, F. Hoffmann-La Roche AG, CH-4070 Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Learnings from previous Roche p38-selective inhibitors were applied to a new fragment hit, which was
optimized to a potent, exquisitely selective preclinical lead with a good pharmacokinetic profile.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 28 February 2011
Revised 23 March 2011
Accepted 24 March 2011
Available online 1 April 2011
Keywords:
p38
Kinase
Kinase inhibitor
The p38 family of MAP kinases, especially the p38a isoform, has
F
F
attracted attention over the last 15 years as an attractive target for
pharmacological intervention in the treatment of inflammatory dis-
eases.1,2 This interest stems from the family’s central role in regu-
H
N
F
F
N
O
O
O
O
N
lating the production of inflammatory cytokines, including TNF-a,
N
OH
IL-1 and IL-6. Biologic agents targeting these cytokines or their
receptors have proven efficacious for inflammatory diseases,
including rheumatoid arthritis, psoriasis and inflammatory bowel
N
N
HN
N
N
HN
O
H2N
3
1
2
HO
OH
disease.3 While the initial excitement over p38
a has been damp-
ened by mediocre phase 2 efficacy results in rheumatoid arthritis
and inflammatory bowel disease for multiple small molecule inhib-
itors, there continues to be interest in this target, for both rheuma-
toid arthritis as well as for pain and respiratory conditions.2c,2d,4
We recently reported on our pyrimidinopyridone inhibitors,
exemplified by 1 (R1487) and 2 (pamapimod), which completed
phase 2 clinical trials for rheumatoid arthritis (Fig. 1).5 As part of
a screening exercise to identify a backup to 2, we discovered inda-
Figure 1. Previously reported Roche p38 inhibitors and screening hit 3.
we expected that the 2,4-difluorophenyl ether sidechain discov-
ered to be optimal in our earlier work would translate well to
the 6-position of indazole-type scaffolds, and allow us to focus
our optimization efforts on other parts of the scaffold. The overlay
also suggested that 3-amino substituents would be well-tolerated.
Pyrazolo[3,4-d]pyrimidines appeared particularly attractive, for
intellectual property and synthetic reasons as well as a perception
that they would have more favorable properties than the corre-
sponding indazoles. We therefore decided to investigate 3-amino-
zole 3 to bind weakly to p38a (Kd = 2 mM as determined by surface
plasmon resonance6). A modeled structure of 3, overlaid onto the
corresponding crystal structure for 1 bound to the ATP-binding site
of p38a, suggested that learnings from our pyridopyrimidine work
would translate well to indazole-type scaffolds (Fig. 2). Specifically,
pyrazolopyrimidines as p38a inhibitors.
Our first generation synthesis is outlined in Scheme 1.7 The
chloride of 2-thiomethyl-4-chloro-5-cyano pyrimidine was dis-
placed with tert-butyl carbazate (affording 4), then the thiomethyl
⇑
Corresponding author. Tel.: +1 973 235 2717; fax: +1 973 235 6263.
E-mail address: michael.soth@roche.com (M. Soth).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.03.098

M. Soth et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3452–3456
3453
F
F
Ar
O
Ar
O
N
(a)
N
(b)
N
Cl
Br
N
Cl
F
F
O
N
N
O
N
O
24
OH
N
25
Br
Ar
Ar
O
Ar
Ar
O
O
O
(c)
(d)
(e)
O
N
O
N
S
HN
HN
S
27
O O
26
Ar
O
Ar
Ar
Ar
N
O
O
N O
N
H
TfO
N
N
BocHN
N
N
S
S
O O
F
O O
F
H
N
N
O
N
N
Figure 2. Modeled structure of screening hit 3 (yellow) overlaid onto a p38a-bound
22
crystal structure of pyrimidinopyridone 1 (orange, PDB accession number 3FLN)
suggests an optimization path for indazole-type scaffolds through 6-(2,4-difluoro-
phenoxy) and 3-amino substitution.
HN
S
O O
Scheme 2. Improved synthesis of 22 (R6226). Detailed conditions can be found in
WO 2007/023110. Reagents and conditions: (a) sodium 2,4-difluorophenoxide,
tetrahydrofuran (70%); (b) (i) iPrMgBr, diethyl ether/tetrahydrofuran, (ii) CO2 (98%);
(c) (i) (COCl)2, cat. N,N-dimethylformamide, dichloromethane, (ii) (R)-2-methane-
sulfonyl-1-methyl-ethylamine, triethylamine, dichloromethane, used crude; (d)
H2O2, HCO2H, dichloromethane (75% from acid); (e) (i) trifluoromethanesulfonic
anhydride, 2,6-lutidine, dichloromethane, (ii) add tert-butyl carbazate, (iii) add
trifluoroacetic acid (64%).
Boc
NH
Cl
N
SMe
HN
NC
N
SMe
(a)
(b)
F
N
N
4
NC
F
Boc
F
H
F
NH
HN
(c)
N
O
N
N
O
Compound 20 was made by a route analogous to that outlined in
Scheme 2. The amide and sulfonamide sidechains were arbitrarily
chosen for synthesis; most of the amine sidechains were selected
based on analogy to examples in our previously published pyrim-
idinopyridone work.5d
N
N
N
NC
6
5
H2N
F
Compounds 6–20 were tested in an enzyme assay for inhibition
of the p38a-catalyzed phosphorylation of myelin basic protein.
F
H
N
(d) or (e)
N
O
Potent inhibitors were identified from both the amide and amine
subseries (Table 1). Amide analogs 7–10 showed relatively shallow
structure–activity relationships (SAR), with a less than 10-fold
range in potencies; the variable portion of the amide subseries is
likely placed out of the binding pocket into solvent and therefore
unlikely to greatly affect binding. The wider range of SAR observed
across the amino analogs 13–22 indicates more stringent
requirements for the amine subseries to optimally fit the protein,
presumably due to closer protein contacts. Most notably, the large
(>100-fold) difference in potencies between example 16 and des-
methyl analog 18 indicates the importance of the pocket filled by
the methyl group (see crystal structure, below). The sulfonamide
subseries was not extensively pursued; while sulfonamide 11
had potency comparable to other compounds shown in Table 1,
it had poor cellular permeability (tested against Caco-2 cells, data
not shown).
Select compounds were also tested for inhibitory activity
against lipopolysaccharide-induced production of interleukin-1 in
human whole blood (HWB). Surprisingly, the upward shifts in
IC50s observed for these compounds between our enzyme assay
and our HWB assay were very small or even reversed (except for
the poorly permeable sulfonamide 11). However, general SAR
trends are consistent; subtle differences between the shifts
N
N
7-19
HN
R
Scheme 1. First generation synthesis of amino pyrazolopyrimidines. Representa-
tive detailed examples can be found in WO 2007/023110. Reagents and conditions:
(a) tert-butyl carbazate, triethylamine, tetrahydrofuran (62%); (b) (i) m-chloroper-
benzoic acid, dichloromethane, (ii) 2,4-difluorophenol, sodium hydride, N,N-
dimethylformamide (47%); (c) (i) trifluoroacetic acid, dichloromethane, (ii) trieth-
ylamine, n-butanol,
D (62%); (d) acid or sulfonyl chloride, pyridine, 1,4-dioxane, RT
or reflux (7–49%); (e) ketone or aldehyde, sodium triacetoxyborohyride, cat. acetic
acid, tetrahydrofuran (20–59%). For compounds 16–18 proceeding through step; (f)
the carbonyl component contained unoxidized sulfides, which were oxidized to the
corresponding sulfones using m-chloroperbenzoic acid or oxone as a final step.
group was oxidized to the corresponding sulfone, which was dis-
placed with 2,4-difluorophenol to afford hydrazide 5. Removal of
the tert-butoxycarbonyl group of 5 released the corresponding
hydrazine, which was then cyclized to afford key intermediate 6.
Intermediate 6 was a suitable starting material for a variety of ana-
logs, and from it we could make amides 7–10 via reaction with acid
chlorides, sulfonamides 11–12 via reaction with the corresponding
sulfonyl chlorides, and amines 13–19 via reductive alkylation.

3454
M. Soth et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3452–3456
Table 1
observed for individual compounds are likely due to differences in
plasma protein binding (data not shown).
Potencies for compounds 6–22
F
Branched sulfone 16 was chosen for further evaluation, based
on its excellent HWB potency as well as a good properties profile,
including high microsomal stability and reasonable solubility.
Enantiomers of 16 were separated by chiral HPLC to afford com-
F
H
N
N
O
N
pounds 21 and 22, and each was co-crystallized with p38a. The
N
crystal structures were used to identify the more potent enantio-
mer as the R-form, 22 (R6226). The higher potency of 22 is easily
HN
R
explainable from its p38a-bound crystal structure ( Fig. 3). The
a
b
Compound
R
p38
a
enzyme IC50
(l
M)
HWB IC50
ND
(lM)
methyl group of 22 fills a small hydrophobic pocket of the protein,
while the sulfone is exposed to solvent. This ideal arrangement is
not available to enantiomer 21.
6
–H
1.00 (0.129)
O
O
7
0.100 (0.046)
ND
Compound 22 also shows extremely high selectivity for p38
over other kinases. Against a 363-kinase panel run by Ambit Biosci-
ences at a test concentration of 10
M,8 22 bound strongly to only
p38 and b. For all other kinases tested, binding was <50%. Kd’s
determined by Ambit Biosciences for 22 against p38 and b were
a
l
8
0.420 (0.100)
0.084 (0.005)
0.446 (0.108)
ND
a
a
O
1 and 25 nM, respectively (10 point concentration curves). We be-
lieve that there are three major factors driving this exquisite
selectivity. The first is an optimal fit of the 2,4-difluorophenyl ether
9
into the back pocket of p38. This pocket is bigger in p38a than in
many other kinases, due to its small gatekeeper residue
O
10
11
0.059 (0.010)
0.257 (0.053)
ND
(Thr106); this design principle has previously been used by us
and others in the design of selective p38a
inhibitors.2c The second
O
S
factor is the presence of a nonplanar front pocket sidechain (the
>10
O
branched sulfone of 22), which is particularly well-tolerated by
p38
a versus other kinases, for example, Lck, due to a larger open
O
space in the front of the protein.2c
12
13
3.58 (0.949)
0.023 (0.006)
ND
S
O
Both of these factors are major contributors to kinase selectiv-
ity. However, they are not sufficient to completely explain the high
kinase selectivity of 22. An instructive comparator is the pyrimidi-
nopyridone analog of 22, compound 23.5d Pyrimidinopyridone 23
0.021 (0.016)
14
15
0.218 (0.69)
ND
is a highly selective p38
a inhibitor and demonstrates the power
O
of the first two selectivity factors. However, it is not as selective
as 22; most notably, it also shows binding to the JNK family of pro-
tein kinases.9
0.038 (0.016)
0.126 (0.002)
N
O
O
S
O
16
17
18
19
0.045 (0.005)
2.20 (1.73)
0.048 (0.014)
S
O
ND
ND
ND
S O
O
O
10.4 (3.32)
S
O
0.690 (0.130)
N
S
O
20
O
O
0.795 (0.204)
ND
O
21
2.34 (0.737)
0.109 (0.025)
ND
S
O
O
22 (R6226)
0.039 (0.006)
S
O
ND = not determined.
a
Inhibition of phosphorylation of myelin basic protein by
. Values are means of three or more experiments. SEM’s are included in
parentheses.
c-
33P-ATP catalyzed by
p38a
b
Inhibition of lipopolysaccharide-stimulated production of interleukin-1 in
human whole blood (HWB). Values are means of two or more experiments. SEM’s
are included in parentheses.
Figure 3. p38a-bound crystal structure of 22 (R6226), PDB accession number
3FMM.

M. Soth et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3452–3456
3455
Table 2
A
satisfying explanation for the improved selectivity of
pyrazolopyrimidine 22 over pyrimidinopyridone 23 comes from
comparison of their p38 -bound crystal structures (Fig. 4). The sul-
In vitro and in vivo properties of compound 22
a
In vitro
fone-containing sidechain sits closer towards the hinge for the pyr-
azolopyrimidine core than it does in the corresponding
pyrimidinopyridone, a consequence of the smaller hinge-binding
motif (five-membered pyrazolo vs six-membered pyrimidino).
Due to this closer fit, the methyl group of the sulfone sidechain
of 22 fits more snugly into a lower hydrophobic pocket defined
in part by Ala157. Ala157 is an uncommon residue in this position,
present in only five kinases, including p38a; for most other kinases
it is a larger residue. The corresponding residue in the JNKs is a va-
line. The tight fit of 22 into this hydrophobic pocket clearly favors
Aq solna
g/mL)
57.2
Cacob
AB/BA
3.0/18.9
Protein
Microsomal
stabilityd
Clint (human) 0.74
Clint (rat) 6.2
(l
bindingc
18.7% free
In vivo rat (Hanover–Wistar)e,f
f
t1/2
Vdss
(L/kg)
8.1
Cl
% F
AUC
Cmax
(h)
3.4
(ml/kg min)
43.6
(ngÃh/mL)
(ng/mL)
446
105
1220
a
Aqueous solubility in phosphate buffer at pH 6.5, resolved by reverse-phase
high performance liquid chromatography (HPLC) with a diode array detector using
an external HPLC standard.
p38
Close inhibitor contact to Ala157 has been noted as a potential
selectivity factor for other p38
inhibitors.10
a over JNK to a greater extent than for the looser fitting 23.
b
Permeability in Caco-2 cells AB (apical to basolateral) and BA (basolateral to
a
apical) movement of 10 lM test compound in 21 days cultured Caco-2 cells (cm/
sec  10EÀ6), pH 7.4 on both sides.
For further development, a chiral and scalable synthesis of 22
was developed (Scheme 2).7,11 Both chlorides of 2,4-dichloro-5-
bromopyrimidine were displaced with 2,4-difluorophenol, and
the resulting bromide 24 was converted to the corresponding orga-
nomagnesium species and reacted with carbon dioxide to afford
acid 25. Acid 25 was condensed with R-2-aminopropyl methyl sul-
fide, and the sulfide of the resulting amide 26 was oxidized to
afford amide 27. The synthesis was completed with a novel one-
pot, three-step sequence in which amide 27 was first treated with
trifluoromethanesulfonic anhydride in the presence of lutidine.
This reaction presumably forms an imidoyl triflate, which was then
treated with tert-butyl carbazate to afford a tert-butoxycarbonyl-
protected hydrazino amidine. The hydrazide could be isolated
but was generally treated directly with trifluoroacetic acid to re-
move the tert-butoxycarbonyl group and allow cyclization of the
resulting hydrazine, with displacement of 2,4-difluorophenol, to
afford 22.
CentrifreeÒ, human plasma.
c
d
Liver microsomal intrinsic clearance (ll/min/mg protein).
Doses were 3.0 mg/kg IV (intravenous) and PO (per os); mean values from three
e
rats per sample time.
Cmax, AUC and % F were determined after the oral dose and Cl, Vdss and t1/2 were
determined from the IV dose.
f
Compound 22 was also a potent inhibitor ex vivo of LPS-induced
cytokine production in rat, blocking formation of TNF-
1.5 mg/kg, EC50 0.76
M).12
In conclusion, a scaffold-hopping exercise from our previously-
reported pyridopyrimidine series of p38 inhibitors yielded a via-
a (ED50
l
a
ble new scaffold, 3-amino pyrazolopyrimidines. A particularly
promising analog, 22, satisfied all internal criteria for selection as
a candidate and was chosen for pre-clinical development.
Modeling: The screening hit 3 was docked into the structure of
Compound 22 showed in vitro properties that translated to
good absorption and high bioavailability in rat (Table 2).
compound 1 bound into p38a
using the program FlexX.13 The top
20 binding modes as prioritized by Screenscore and FlexX scoring
functions were visually examined. The selected binding mode
was optimized in the protein environment using the MAB force-
field14 and 500 steps of conjugate gradient. No constraints were
placed on the ligand atoms during minimization. Binding pocket
residues within 5 Å of ligand atoms were kept free during optimi-
zation while the rest of the protein atoms were fixed. Binding
modes have been captured in the manuscript using the program
PyMOL.15
Protein crystallography: X-ray crystal structures of p38a-ligand
complexes were determined following published procedures.16
Atomic co-ordinates and structure factor amplitudes have been
deposited to the RCSB PDB along with detailed statistics on data
collection and structure refinement.
References and notes
1. (a) Schett, G.; Zwerina, J.; Firestein, G. Ann. Rheum. Dis. 2008, 67, 909.
2. (a) Goldstein, D. M.; Gabriel, T. Curr. Top. Med. Chem. 2005, 5, 1017; (b) Westra,
J.; Limburg, P. C. Mini-Rev. Med. Chem. 2006, 6, 867; (c) Goldstein, D. M.;
Kuglstatter, A.; Lou, Y.; Soth, M. J. J. Med. Chem. 2010, 53, 2345; (d) Genovese, M.
C.; Gao, L.; Yin, J.; Smith, S.; Weinblatt, M.; Smolen, J.; Wang, X.; Schieven, G.;
Garcia-Meijide, J. A.; Latek, R.; Pasternak, R.; Kaul, S.; Roy, A.; Raymond, R.,
Thienel, U.; Wang, J. Presented at the American College of Rheumatology
Scientific Meeting; Atlanta, GA, 2010; Poster 1119.
3. Schett, G.; Stach, C.; Zwerina, J.; Voll, R.; Manger, B. Arthritis Rheum. 2008, 58,
2936.
4. Genovese, M. C. Arthritis Rheum. 2009, 602, 317; Hammaker, D.; Firestein, G. S.
Ann. Rheum. Dis. 2010, 69, 77 (Suppl. I.
5. (a) Hill, R. J.; Dabbagh, K.; Phippard, D.; Li, C.; Suttmann, R. T.; Welch, M.; Papp,
E.; Song, K. W.; Chang, K.-C.; Leaffer, D.; Kim, Y.-N.; Roberts, R. T.; Zabka, T. S.;
Aud, D.; Dal Porto, J.; Manning, A. M.; Peng, S. L.; Goldstein, D. M.; Wong, B. R. J.
Pharmacol. Exp. Ther. 2008, 327, 610; (b) Cohen, S. B.; Cheng, T. T.; Chindalore,
V.; Damjanov, N.; Burgos-Vargas, R.; Delora, P.; Zimany, K.; Travers, H.;
Caulfield, J. P. Arthritis Rheum. 2009, 60, 335; (c) Alten, R. E.; Zerbini, C.; Jeka, S.;
Irazoque, F.; Khatib, F.; Emery, P.; Bertasso, A.; Rabbia, M.; Caulfield, J. P. Ann.
Figure 4. Connolly surface on p38a-bound crystal structures of 22 (PDB accession
number 3FMM) and 23 (3FLS) illustrates that the methyl group of 22 is positioned
to make better contacts with the small pocket defined by Ala157.

3456
M. Soth et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3452–3456
Rheum. Dis. 2010, 69, 364; (d) Goldstein, D. M.; Soth, M. J.; Gabriel, T.; Dewdney,
11. Analytical data for 22 (R6226): Mp 176.9–178.8. 1H NMR (400 MHz, DMSO-d6) d
ppm 1.40 (d, J = 6.57 Hz, 3H) 3.01 (s, 3H) 3.24 (dd, J = 14.15, 6.57 Hz, 1H) 3.56
(dd, J = 14.15, 5.56 Hz, 1H) 4.21 (dt, J = 13.26, 6.76 Hz, 1H) 6.88 (d, J = 8.08 Hz,
1H) 7.07–7.23 (m, 1H) 7.35–7.55 (m, 2H) 8.93 (s, 1H) 12.48 (s, 1H). MS (EI/CI)
m/z: (M+H) 384. Anal. (C15H15F2N5O3S) C, H, N. Specific rotation (MeOH)-
34.000 °/g/100 mL.
N.; Kuglstatter, A.; Arzeno, H.; Chen, J.; Bingenheimer, W.; Dalrymple, S. A.;
Dunn, J.; Farrell, R.; Frauchiger, S.; La Fargue, J.; Ghate, M.; Graves, B.; Hill, R. J.;
Li, F.; Litman, R.; Loe, B.; McIntosh, J.; McWeeney, D.; Papp, E.; Park, J.; Reese, H.
F.; Roberts, R. T.; Rotstein, D.; San Pablo, B.; Sarma, K.; Stahl, M.; Sung, M.;
Suttman, R. T.; Sjogren, E. B.; Tan, Y.; Trejo, A.; Welch, M.; Weller, P.; Wong, B.
R.; Zecic, H. J. Med. Chem. 2011, 54, 2255.
12. All animal procedures were approved by the Institutional Animal Care and Use
Committee of Roche Palo Alto. Compound was administered (p.o.) to Hanover–
6. Huber, W.; Mueller, F. Curr. Pharm. Des. 2006, 12, 3999.
7. Representative experimental procedures for compounds can be found in WO
2007/023110.
Wistar rats 0.5 h prior to challenge with 50 lg/kg LPS (i.e., 0.9% saline). Serum
was collected 1.5 h after LPS injection and analyzed for TNF-
and drug concentrations.
a levels (ELISA)
8. Fabian, M. A.; Biggs, W. H., III; Treiber, D. K.; Atteridge, C. E.; Azimioara, M. D.;
Benedetti, M. G.; Carter, T. A.; Ciceri, P.; Edeen, P. T.; Floyd, M.; Ford, J. M.;
Galvin, M.; Gerlach, J. L.; Grotzfeld, R. M.; Herrgard, S.; Insko, D. E.; Insko, M. A.;
Lai, A. G.; Lelias, J.-M.; Mehta, S. A.; Milanov, Z. V.; Velasco, A. M.; Wodicka, L.
M.; Patel, H. K.; Zarrinkar, P. P.; Lockhart, D. J. Nat. Biotechnol. 2005, 23, 329.
13. (a) Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. J. Mol. Biol. 1996, 261, 470; (b)
Rarey, M.; Kramer, B.; Lengauer, T. J. Comput. Aided Mol. Des. 1997, 11, 369.
14. (a) Gerber, P. R.; Muller, K. J. Comput. Aided Mol. Des. 1995, 9, 251; (b) Gerber, P.
R. J. Comput. Aided Mol. Des. 1998, 12, 37.
15. DeLano, W. L. The PyMOL Molecular Graphics System; DeLano Scientific: Palo
Alto, CA, USA, 2002. http://www.pymol.org.
9. Out of 127 kinases tested at Ambit Biosciences,8 compound 23 at 10
lM
displaces ligands for the following kinases at the percents indicated: p38
a,
p38b, JNK1, JNK2, and JNK3 (100%); CSNK1E (88%); GAK (87%); and STK36
16. Trejo, A.; Arzeno, Humberto; Browner, Michelle F.; Chanda, Sushmita; Cheng,
Soan; Comer, Daniel D.; Dalrymple, Stacie A.; Dunten, Pete; La Fargue, JoAnn;
Lovejoy, Brett; Freire-Moar, Jose; Lim, Julie; McIntosh, Joel; Miller, Jennifer;
Papp, Eva; Reuter, Deborah; Roberts, Rick; Sanpablo, Florentino; Saunders,
John; Song, Kyung; Villasenor, Armando; Warren, Stephen D.; Welch, Mary;
Weller, Paul; Whiteley, Phyllis E.; Zeng, Lu; Goldstein, David M. J. Med. Chem.
2003, 46, 4702.
(74%). Ambit Kd’s were determined for p38a (1.2 nM), p38b (8.7 nM), JNK1
(110 nM), JNK2 (16 nM) and JNK3 (16 nM).
10. Herberich, B.; Cao, G-Q.; Chakrabarti, P. P.; Falsey, J. R.; Pettus, L.; Rzasa, R. M.;
Reed, A. B.; Reichelt, A.; Sham, K.; Thaman, M.; Wurz, R. P.; Xu, S.; Zhang, D.;
Hsieh, F.; Lee, M. R.; Syed, R.; Li, V.; Grosfeld, D.; Plant, M. H.; Henkle, B.;
Sherman, L.; Middleton, S.; Wong, L. M.; Tasker, A. S. J. Med. Chem. 2008, 51,
6271.