Pyrazinoindolone inhibitors of MAPKAP-K2

Article abstract of DOI:10.1016/j.bmcl.2007.12.037

Optimization of pyrazinoindolone inhibitors of MAPKAP-K2 (MK2) provides a reasonable balance of cellular potency and physicochemical properties. Mechanistic studies support the inhibition of MK2 which is responsible for the sub-micromolar cellular efficacy.

Full text of DOI:10.1016/j.bmcl.2007.12.037

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 938–941
Pyrazinoindolone inhibitors of MAPKAP-K2
D. R. Goldberg,a,* Y. Choi,a D. Cogan,a M. Corson,d R. DeLeon,c A. Gao,a
L. Gruenbaum,c M. H. Hao,a D. Joseph,d M. A. Kashem,a C. Miller,a N. Moss,a
M. R. Netherton,a C. P. Pargellis,b J. Pelletier,b R. Sellati,b D. Skow,a C. Torcellini,d
Y.-C. Tseng,d J. Wang,a R. Wasti,c B. Werneburg,b J. P. Wua and Z. Xionga
aDepartment of Medicinal Chemistry, Boehringer Ingelheim Pharmaceuticals, Inc., Research and Development Center,
900 Ridgebury Road, Ridgefield, CT 06877, USA
bDepartment of Immunology and Inflammation, Boehringer Ingelheim Pharmaceuticals, Inc., Research and Development Center,
900 Ridgebury Road, Ridgefield, CT 06877, USA
cDepartment of Translational Sciences, Boehringer Ingelheim Pharmaceuticals, Inc., Research and Development Center,
900 Ridgebury Road, Ridgefield, CT 06877, USA
dDepartment of Drug Discovery Support, Boehringer Ingelheim Pharmaceuticals, Inc., Research and Development Center,
900 Ridgebury Road, Ridgefield, CT 06877, USA
Received 2 December 2007; revised 16 December 2007; accepted 17 December 2007
Available online 23 December 2007
Abstract—Optimization of pyrazinoindolone inhibitors of MAPKAP-K2 (MK2) provides a reasonable balance of cellular potency
and physicochemical properties. Mechanistic studies support the inhibition of MK2 which is responsible for the sub-micromolar
cellular efficacy.
ꢀ 2007 Elsevier Ltd. All rights reserved.
The mitogen-activated protein kinases (MAPKs) are
important components of signal transduction pathways
that mediate cell proliferation, differentiation, and
death.1 The p38 MAPK family has been shown to regu-
late the production of a number of pro-inflammatory
cytokines such as TNFa and IL-1b, both of which are
implicated in chronic inflammatory diseases.2 MAP-
KAP-K2 (MK2), a direct substrate of p38, has been
shown to play an important role in TNFa production
in mice genetically deficient in MK2.3–5 Lipopolysaccha-
ride (LPS) stimulated MK2 knockout mice produce a
significantly reduced amount of TNFa compared to
wild type animals.3 In addition, MK2 knock-out mice
are resistant to developing disease in arthritis models.4
Furthermore, MK2 has been shown to have fewer sub-
strates compared to p38 which may provide for a poten-
tially safer target.5–7
Recently we disclosed a class of carboline based MK2
inhibitors exemplified by compounds 1 and 3 (scaffold
A, Table 1).8 However, only analog 1 had potency in
an assay of LPS-induced THF production in THP-1
cells. This report details efforts to improve the cellular
potency of our MK2 inhibitors while balancing suitable
physicochemical properties.
We found that transposing the indole nitrogen from the
carboline scaffold to the corresponding pyrazinoindo-
lone scaffold (scaffold B, compounds 2 vs 1 and 4 vs 3,
Table 1) provided us with a small improvement in
molecular potency. We found this potency increase to
be general and the same SAR trends were observed for
compounds containing either the carboline or pyrazino-
indolone scaffold.8 Given that previous SAR on the thi-
azole carboxamide (R in A) provided us with our largest
improvement in molecular potency, and that the pyra-
zinoindole scaffold was more potent than the carboline
scaffold, we focused on improving cell potency by mod-
ifying the R-group on the pyrazinoindolone, scaffold B.
Keywords: MAPKAP-K2; THP-1.
*
Compounds in Tables 1–3 were prepared according to
Scheme 1. 4-Methyl-3-nitrobenzoic acid (5) was first
Corresponding author. Tel.: +1 203 778 7828; fax: +1 203 791
6072; e-mail: dgoldber@rdg.boehringer-ingelheim.com
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.12.037

D. R. Goldberg et al. / Bioorg. Med. Chem. Lett. 18 (2008) 938–941
939
Table 1. Isomeric indole
Table 3. Indolonone SAR
H
O
N
O
H
N
O
H
N
N
N
O
NH
N
N
NH
1
R
R
S
O
N
2
R
N
H
S
O
R
H
A
Compound
R
R1/R2
MK2
THP-1
O
H
N
a
IC50 (nM) IC50 (nM)
a
O
N
NH
R
N
S
O
H
21
28
CH3/H 24
CH3/CH3 19
1600
1000
B
a
Compound
Scaffold
R
MK2 IC50
(nM)
THP-1
a
IC50 (nM)
4
29
30
CH3/H
CH3/CH3
(CH2)3
5
5
2
>5000
2400
1700
N
1
2
A
B
44
20
1600
1400
3
4
A
B
10
5
>5000
>5000
N
26
31
32
CH3/H
CH3/CH3
(CH2)3
3
2
2
735
430
300
N
a Values are means of P2 experiments, standard deviation typically
50% of reported value.
27
33
34
CH3/H
CH3/CH3
(CH2)3
9
5
3
1,600
900
520
protected as the corresponding ethyl ester and then
reacted with diethyloxalate under basic conditions to
provide the corresponding aryl pyruvate ester 6. Indole
formation occurred upon heating in zinc and acetic acid
to afford 7.9 Alkylation of the indole nitrogen with
either bromoacetonitrile or 2-bromopropionitrile pro-
vided 8 and 9, respectively. Compound 8 was alkylated
with either methyl iodide or 1,3-dibromopropane to 10
or 11.10 Subjecting 9–11 to reductive conditions with
either CoCl2/NaBH4 or H2 over platinum (IV) oxide
produced lactams 12–14.11 Amide bond formation with
N
a Values are means of P2 experiments, standard deviation typically
50% of reported value.
EDC and 2-aminothiazole-4-carboxylate followed by
ester hydrolysis gave 15–17. Finally, amide analogs 18–
20 were prepared via EDC mediated coupling of the cor-
responding thiazole acids with a variety of amines.
Table 2. Thiazole carboxamide SAR
O
H
N
O
N
N
NH
R
S
O
N
H
a
a
THP-1 IC50 (nM)
Compound
R
MK2 IC50 (nM)
5
PAMPA (pE ·10ꢀ6
)
4
>5000
0
N
2
20
12
1400
5800
9
21
Not tested
O
22
23
24
25
26
27
45b
46
7
1800b
2700
8900
1670
735
Not tested
8.6
0.1
N
4
4.9
N
N
3
20.8
5.0
9
1600
N
a Values are means of P2 experiments, standard deviation typically 50% of reported value.
b Values n = 1.

940
D. R. Goldberg et al. / Bioorg. Med. Chem. Lett. 18 (2008) 938–941
O
O
O
O
a
b
HO
O
NO2
O
N
H
O
NO2
O
R1 = R2 = H
O
O
Br
R1
CN
R2
5
6
7
R1= H; R2 = CH3
c
O
O
f or g
O
N
O
NH
N
O
R1
R1
O
R2
O
CN
R2
12: R1 = H; R2 = CH3
13: R1 = R2 = CH3
14: R1 = R2 = (CH2)2
8: R1 = R2 = H
9: R1 = H; R2 = CH3
h, i
d
e
10: R1 = R2 = CH3
11: R1 = R2 = (CH2)2
O
O
H
H
N
O
N
N
O
j
N
N
NH
N
NH
R1
R1
O
S
HO
R2
R
O
S
N
R2
H
15: R1 = H; R2 = CH3
16: R1 = R2 = CH3
17: R1 = R2 = (CH2)2
18: R1 = H; R2 = CH3
19: R1 = R2 = CH3
20: R1 = R2 = (CH2)2
Scheme 1. Reagents and conditions: (a) 1—HCl (g), EtOH, 90 ꢁC, 98%; 2— NaOEt, EtOH, diethyloxalate, rt, 77%; (b) Zn, HOAc–H2O, 75 ꢁC, 81%;
(c) K2CO3, DMF, 80 ꢁC, 88–91%; (d) LiHMDS, CH3I, THF, 0 ꢁC–rt, 84–95%; (e) 1,3-dibromopropane, 60% NaH, DMSO-ether, 65 ꢁC, 61–95%; (f)
PtO2, 50 psi H2, EtOH, 54–95%; (g) CoCl2, NaBH4, MeOH–THF, 70 ꢁC, 59%; (h) Methyl-2-aminothiazole-4-carboxylate EDC, HObT, DMAP
(cat.), DMF, rt–60 ꢁC, 53–94%; (i) NaOH, H2O–MeOH, 60 ꢁC, 93–98%; (j) RNH2, EDC, HObT, DMAP (cat.), DMF, rt–60 ꢁC, 25–90%.
Table 2 highlights a number of amide R-groups on the
thiazole with scaffold B. While analogs of our 4 that
lacked the piperidene nitrogen (2 and 21–23) consis-
tently had measurable cell potency, these compounds
had poor physicochemical properties (solubility at pH
7.4 61.5 lg/mL vs >96 lg/mL for compound 4). Given
that the piperidene had reasonable physicochemical
properties, but not cellular efficacy, we wanted to deter-
mine if we could build from the piperidine nitrogen in
order to provide a balance of cellular potency and suit-
able physicochemical properties. We found that the
addition of lipophilic alkyl or aryl groups on the pipe-
ridene nitrogen (24–27) allowed us to maintain reason-
able physicochemical properties (pH 7.4 solubility
generally >30 lg/mL) while providing THP-1 activity.
While a number of factors likely govern cell efficacy,
we found that compounds which demonstrated better
permeability in the parallel artificial membrane perme-
ability assay (PAMPA) tended to provide better cellular
efficacy versus 4.12,13
While 32 shows negligible inhibition of the following ki-
nases at 10 lM: Erk2, ANKRD3, CDK2, smLck,
MKK1, PTK2, prkCDK, NEK7, MAPK3, CK1,
PRK2, MST4, ITK, MAPKAP-K3, p38-a, -b, -c, -d,
AURORAb, PLK1, Lyn, NEK6, MSKk1, MAP3K11,
CHK1, NEK2A, MNK1, CHK2, PDK1, PKCa,
PRKc1, CK2, PKA, PIM2, MBK5, SGK, DAPK1,
PLKK, CSK, EF2K, VEGFR, PKBdph, Src, PKBb,
Jnk1a1, Jnk3a1; the kinases listed in Table 4 were all
inhibited to a significant extent.
Given the kinase selectivity profile of 32, we wanted to
confirm that the potency in the THP-1 cell assay was pri-
marily due to MK2 inhibition. We therefore tested both
Table 4. Kinase selectivity of 32
O
NH
H
O
N
N
N
N
S
O
N
H
From exploring SAR at other regions of 4, we found
that substitution of the 4-position of the indole with
either a gem-dimethyl or spirocyclobutyl group provided
small increases in both molecular and cellular potency
without affecting our improved PAMPA permeability
(Table 3). Overall, compound 32 provided us our best
combination of molecular, cellular, and physicochemical
properties and we therefore profiled this compound fur-
ther to assess the potential for this class of MK2
inhibitors.
Kinase
IC50 (nM)
(%inh. 10 lM)
Kinase IC50 (nM)
(%inh. 10 lM)
MAPKAP-K1
Erk2
PBK
25
NIK
PDK1
480
>3000
(75)
(94)
(82)
(86)
93
(92)
(84)
(98)
(60)
(63)
920
(80)
CHK2
PRAK
SRPK1
ROCKII
GSK3b
MNK2a
DYRK1A
ERK8
MST2
CLK1
AMPK
(98)

D. R. Goldberg et al. / Bioorg. Med. Chem. Lett. 18 (2008) 938–941
941
Table 5. Cellular efficacy studies of 32a
Overall, we have been able to improve the profile of our
initially disclosed carboline based MK2 inhibitors. We
have demonstrated a better balance of physicochemical
properties and cellular efficacy. Furthermore, we have
demonstrated this potency is due to inhibition of MK2
via examining both upstream and downstream effects
of our inhibitors in the MK2 pathway.
O
H
O
N
N
N
NH
N
S
O
N
H
MK2 IC50 (nM)
2
LPS stimulation of THP-1 cells IC50 (nM)
TNFa
300
Anisomycin-stimulated HeLa cells IC50 (nM)
P-p38a
References and notes
>30,000
710
1. Imajo, M.; Tsuchiya, Y.; Nishida, E. IUBMB Life 2006,
58, 312.
P-HSP27 cellular efficacy IC50 (nM)
THP-1
2. (a) Wagner, G.; Laufer, S. Med. Res. Rev. 2005, 26, 1; (b)
Calcagni, F.; Elenkov, I. Ann. N.Y. Acad. Sci. 2006, 62.
3. Kotlyarov, A.; Neininger, A.; Schubert, C.; Eckert, R.;
Birchmeier, C.; Volk, H.-D.; Gaestel, M. Nat. Cell Biol.
1999, 1, 94.
4. Winzen, R.; Kracht, M.; Ritter, B.; Wilhelm, A.; Chen, C.-
Y. A.; Shyu, A.-B.; Muller, M.; Gaestel, M.; Resch, K.;
Holtmann, H. EMBO J. 1999, 18, 4969.
5. Gaestel, M. Nat. Rev. Mol. Cell Biol. 2006, 7, 120.
6. Anderson, D. R.; Mahoney, M. W.; Phillion, D. P.;
Rogers, T. E.; Meyers, M. J.; Poda, G.; Hegde, S. G.;
Singh, M.; Reitz, D. B.; Wu, K. K.; Buchler, I. P.; Xie, J.;
Vernier, W. F. WO 2004/058762 A1, 2004.
a Values are means of P2 experiments, standard deviation typically
50% of reported value.
Table 6. BalbC mouse pharmacokinetics
Compound
1
2
32
IV dose 0.5 mg/kg (70:30 PEG, 400:H2O)
18
Cl (%Q)
MRT (h)
18
2.0 1.0
535 165
30
6.1 0.7
165 34
0.8 0.07
525 140
AUC0–1 (ng h/mL)
PO dose 1.0 mg/kg (PEG 400)
Cmax (lM)
AUC0–1 (ng h/mL)
%F
0.06 0.01
88 1.8
0.2
0.08 0.05
115 17
0.06 0.01
446 56
>100 17
7. (a) Rousseau, S.; Peggie, M.; Campbell, D. G.; Nebreda,
A. R.; Cohen, P. Biochem. J. 2005, 391, 433; (b) Adams,
R. H.; Porras, A.; Alonso, G.; Jones, M.; Vintersten, K.;
Panelli, S.; Valladares, A.; Perez, L.; Klein, R.; Nebreda,
A. R. Mol. Cell 2000, 6, 109.
8
11
2
the upstream and downstream effects of our inhibitors in
the p38 MAPK pathway. This was assessed in two dif-
ferent ways (Table 5). First, inhibitor treated HeLa cells
that were stimulated with anisomycin to maximally acti-
vate the p38/MK2 pathway did not show any inhibition
of p38 phosphorylation by flow cytometry ruling out
compound effects on signaling events upstream of
MK2. Second, 32 inhibited phosphorylation of HSP27,
a direct downstream substrate of MK2, to a similar level
of potency to that observed in the THP-1 cell assay.13,14
Taken together, this supports that the cellular efficacy
we are observing is driven by MK2 inhibition.
8. Wu, J.-P.; Wang, J.; Abeywardane, A.; Andersen, D.;
Emmanuel, M.; Gautschi, E.; Goldberg, D. R.; Kas-
hem, M. A.; Lukas, S.; Mao, W.; Martin, L.; Morwick,
T.; Moss, N.; Pargellis, C.; Patel, U. R.; Patnaude, L.;
Peet, G. W.; Skow, D.; Snow, R. J.; Ward, Y.;
Werneburg, B.; White, A. Bioorg. Med. Chem. Lett.
2007, 17, 46649.
9. Kermack, W. O. J. Chem. Soc., Trans. 1924, 2285.
10. Bellemin, R.; Decerprit, J.; Festal, D. Eur. J. Med. Chem.
1996, 31, 123.
11. Heinzman, S. W.; Ganem, B. J. Am. Chem. Soc. 1982, 104,
6801.
12. Levin, J. I.; Du, M. T. Drug Des. Discov. 2003, 18, 123.
13. Anderson et al., hypothesize in their MK2 paper that a
potential reason for a large shift between molecular and
cellular potency may be due to the affinity of MK2 for
ATP in which they calculated a theoretical shift in potency
based on the measured in vitro Km of ATP for MK2 to be
30 lM, see: Anderson, D. R.; Meyers, M. J.; Vernier, W.
F.; Mahoney, M. W.; Kurumbail, R. G.; Caspers, N.;
Poda, G. I.; Schindler, J. F.; Reitz, D. B.; Mourey, R. J. J.
Med. Chem. 2007, 50, 2647, and references therein.
14. Stokoe, D.; Engel, K.; Campbell, D. G.; Cohen, P.;
Gaestel, M. FEBS Lett. 1992, 313, 307.
15. Goldberg, D. R.; Hao, M. H.; Qian, K. C.; Swinamer, A.
D.; Gao, D. A.; Xiong, Z.; Sarko, C.; Berry, A.; Lord, J.;
Magolda, R. L.; Fadra, T.; Kroe, R. R.; Kukulka, A.;
Madwed, J. B.; Martin, L.; Pargellis, C.; Skow, D.; Song,
J. J.; Tan, Z.; Torcellini, C. A.; Zimmitti, C. S.; Yee, N.
K.; Moss, N. J. Med. Chem. 2007, 50, 4016.
The mouse pharmacokinetic properties for this class of
compounds are described in Table 6 and indicate that
overall, 32 is better than either compounds 1 and 2. This
may in part be again due to the poor aqueous solubility
of 1 and 2 (pH 7.4 <0.1 lg/mL) while 32 is considerably
more soluble (pH 7.4 >96 lg/mL).
In order to assess the activity of our compounds in vivo,
a close analog of 32 (compound 26) was tested in a
mouse model of LPS-stimulated TNFa production.15
However, even at an oral dose of 60 mg/kg, we did not
observe any inhibition of TNFa. This is not surprising
given the low plasma levels which are barely above the
THP-1 cell potency (90 min post-dose = 690 nM; THP-
1 IC50 = 735 nM).