3,5-Disubstituted-indole-7-carboxamides: The discovery of a novel series of potent, selective inhibitors of IKK-β

Article abstract of DOI:10.1016/j.bmcl.2011.02.107

The discovery and hit-to-lead exploration of a novel series of selective IKK-β kinase inhibitors is described. The initial lead fragment 3 was identified by pharmacophore-directed virtual screening. Homology model-driven SAR exploration of the template le

Full text of DOI:10.1016/j.bmcl.2011.02.107

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2255–2258
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
3,5-Disubstituted-indole-7-carboxamides: The discovery of a novel series
of potent, selective inhibitors of IKK-b
a
a,
⇑
a
a
a
David D. Miller , Paul Bamborough , John A. Christopher , Ian R. Baldwin , Aurelie C. Champigny ,
a
b
a
a
a
Geoffrey J. Cutler , Jeffrey K. Kerns , Timothy Longstaff , Geoffrey W. Mellor , James V. Morey ,
a
b
b
b
Mary A. Morse , Hong Nie , William L. Rumsey , John J. Taggart
a
GlaxoSmithKline R&D, Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire SG1 2NY, UK
GlaxoSmithKline Inc., 709 Swedeland Road, King of Prussia, PA 19406, USA
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The discovery and hit-to-lead exploration of a novel series of selective IKK-b kinase inhibitors is
described. The initial lead fragment 3 was identified by pharmacophore-directed virtual screening.
Homology model-driven SAR exploration of the template led to potent inhibitors, such as 12, which dem-
onstrate efficacy in cellular assays and possess encouraging developability profiles.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 2 February 2011
Revised 24 February 2011
Accepted 25 February 2011
Available online 1 March 2011
Keywords:
Kinase
IKK-b
IKK
Indolecarboxamide
Protein kinases are regulators of a broad range of cellular pro-
cesses. Consequently, kinases involved in disease-associated aber-
rant signalling are considered attractive targets for drug discovery.
In vitro potency
Selectivity
Cellular potency
pIC50 > 6.5
>10 fold over IKK-α
pIC50 > 5.7 in relevant cells
I
j
B kinase b (IKK-b or IKK-2) is a serine/threonine kinase located
on an isolated branch of the kinase phylogenetic tree along with
homologous family members, IKK- (IKK-1), IKK- (IKK-3) and
Figure 1. Minimum target lead profile; pIC50 = ꢀ(log10 IC50).
3
a
e
1
TBK1 (IKK-4). Ubiquitously expressed, IKK-b is a member of the
IKK signalsome which regulates the signalling pathway that links
cellular engagement of pro-inflammatory stimuli such as lipopoly-
saccharide (LPS) and tumour necrosis factor-
scription factor Nuclear Factor kappa B (NF-
by IKK-b of I B (inhibitor of kappa B) in a cytosolic complex with
searches. The objective in the lead discovery phase was to find no-
vel IKK-b inhibitor templates that would meet the minimum target
profile described in Figure 1 and, in addition, have suitable drug-
like properties to provide confidence to initiate a lead optimization
a
(TNF-
a
) to the tran-
2
jB). Phosphorylation
j
programme. Although IKK-a (IKK-1), which is also found in the
NF-
of I
j
B leads to the ubiquitination and proteosomal degradation
IKK signalsome, is reported to be much less active as a kinase than
9
jB. This releases NF-kB and allows it to translocate to the nu-
IKK-b, it has been reported to have non-redundant roles in signal-
1
0
cleus, where it regulates the expression of multiple genes, leading
ling and so IKK-b inhibitors with selectivity over IKK-a were
to a range of immune and inflammatory responses.2 IKK-b inhibi-
,3
sought as a complement to dual-active compounds.
11
tion alone is believed to be sufficient to block the activation of NF-
In the absence of crystallographic data on IKK-b, a 3D pharma-
cophore was built from a homology model based on the X-ray
structure of CDK2 complexed with 1, a close analogue of a litera-
ture thiophene carboxamide IKK-b inhibitor TPCA-1 (Fig. 2).1
A 3D pharmacophore search of the GSK screening collection
(Fig. 2) identified several hundred compounds which were tested
for IKK-b inhibition. From this, a small number of weakly active
hits containing the motifs shown in Figure 3 were identified. These
results contributed to the observation that a variety of IKK-b inhib-
itors contain a primary carboxamide held in a constrained confor-
mation by an adjacent hydrogen-bonding functionality. A separate
4
j
B
and as a result the search for small-molecule inhibitors of
B-associated diseases
such as asthma, COPD and rheumatoid arthritis has attracted much
IKK- b as drugs for use in the control of NF-
j
2,13
5
–8
attention.
Here, we describe the discovery and preliminary SAR of a series
of indole-7-carboxamide IKK-b inhibitors that were identified
through the application of pharmacophore-directed database
⇑
Corresponding author. Tel.: +44 1438 763246.
E-mail address: paul.a.bamborough@gsk.com (P. Bamborough).
0
960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.02.107

2
256
D. D. Miller et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2255–2258
O
OMe
O
OH
Boc
H
H
N
(a) - (c)
N
(d) - (g)
N
O
NH2
Br
H
N
4
5
6
H N
2
S
O
TPCA-1
O
N(Boc)2
Br
O
NH2
F
Boc
H
O
NH2
(h), (i)
(j), (k)
N
N
H
N
Ar
H N
2
S
O
7
8
1
Scheme 1. Reagents and conditions: (a) Boc
ꢀ78 °C; (c) ClC(O)OMe, ꢀ78 °C; (d) NBS, CH Cl
g) LiOH, MeOH, ; (h) NH /dioxane, HATU, DIPEA, CH
CH Cl rt; (j) ArB(OH) , K CO , PdCl (dppf), CH Cl , DMF, 80 °C; (k) cHCl, EtOH,
-wave, 150 °C, 5 min.
2
O, THF, rt; (b) s-BuLi, TMEDA, THF,
, rt; (e) TFA, rt; (f) MnO , THF, rt;
Cl rt; (i) Boc O, DMAP,
Et
2
2
2
(
D
3
2
2
2
Figure 2. IKK-b pharmacophore based on X-ray complex between thiophene
carboxamide 1 (IKK-b pIC50 8.1) and CDK2. Aromatic centroid (orange) and H-bond
donor (magenta) and acceptor (cyan) vectors are shown.
2
2
2
2
3
2
2
2
l
series of palladium-catalysed arylation reactions followed by
deprotection to the target compounds 8.
Preliminary exploration of the SAR around the template indi-
cated that a variety of groups was tolerated at the 5-position of
the indole (Table 1). With a substituted phenyl group in this posi-
tion, potency at IKK-b generally increased in the order ortho
O
NH2
H N
O
NH2
2
H N
2
Figure 3. Frequently occurring motifs observed in preliminary hits.
<
a
meta < para (e.g., 8c–8e) and the promising selectivity over IKK-
improved between meta and para (cf. 8d with 8e and 8f with
exercise subsequently exploited this to find inhibitors of both
8
g). Early indications suggested that heterocycles could also be
IKK-a
and IKK-b kinases.11
accommodated in this position (8h–8j). These results provided
confidence that there would be significant scope for modulation
of potency and compound properties through variation at the
A substructure search based on the these motifs led to the test-
ing in the IKK-b enzyme assay of tricyclic indole derivative 2
4
(
Fig. 4), which was found to be modestly active.1 Compound 2
5
-position of the indole during lead optimisation.
1
5
resembles another series of b-carboline IKK-b inhibitors. Since
the b-carbolines do not contain the primary carboxamide motif
essential for the activity of the indole carboxamides (see later,
Fig. 6) we believe them to be distinct.
Docking of structure 3 into the IKK-b homology model pre-
dicted a binding mode that is consistent with that of thiophene 1
in the X-ray complex with CDK-2. The carboxamide forms hydro-
gen-bonds to Glu97 and Cys99 of the kinase hinge region and the
indole NH holds the carboxamide co-planar with the indole via a
SAR knowledge from the thiophene series (e.g. TPCA-1) sug-
gested that the cyclohexene ring could be removed and that the
bromine in the 5-position could be replaced by other groups such
as aryl rings. Carrying out these changes did indeed produce a lead
compound 3 that retained potency at IKK-b in the micromolar
range. 3 has high ligand efficiency, low molecular weight and
low lipophilicity (LE = 0.44, MW = 236, c Log P = 2.7).16
13
further intramolecular hydrogen-bond (Fig. 5).
The binding hypothesis was supported by the observation that
compounds 9 and 10, which were made by variation of the route
described in Scheme 1 and which were expected to be unable to
bind in the predicted manner, were both undetectable in the
IKK-b assay (Fig. 6).
A further attraction of the indole core as a potential lead tem-
plate was the access it offered from its 3-position to regions of
the IKK-b active site that were not accessible from the correspond-
ing position of thiophene-based inhibitors such as 1 because of the
position of the sulphur atom. The key protein residues that were
The route devised for the synthesis of 3 and related analogues is
1
7
described in Scheme 1. Boc protection of indoline 4 followed by
lithiation and quenching with methyl chloroformate provided the
protected, carboxylated indoline 5 which was brominated, Boc-
deprotected, oxidised to the indole and the ester saponified to give
the key, versatile intermediate 6. Conversion to the primary amide
and di-protection yielded 7, which was used as a substrate in a
Table 1
IKK-b and IKK-
a
inhibition by compounds 3 and 8a–j14
O
NH2
O
NH
2
IKK-b pIC50a
pIC50a
IKK-a
Compounds
Ar
H
H
N
N
3
Ph
3-CN-Ph
5.8
5.8
6.3
5.2
6.1
6.5
5.6
5.9
5.3
5.8
6.3
5.0
5.0
5.4
<4.8
5.3
5.2
4.8
4.8
4.8
4.9
5.4
8a
O
NH2
Br
8b
8c
4-MeSO
2-Cl-Ph
3-Cl-Ph
4-Cl-Ph
3-HOCH
4-HOCH
5-Pyrimidinyl
4-Methoxy-3-Pyridinyl
2-Benzofuranyl
2
NH-Ph
H
N
8d
2
H N
2
3
S
8e
8f
O
IKK-β pIC = 5.6
IKK-β pIC = 5.8
2
-Ph
-Ph
5
0
50
8
8
8
g
h
i
2
TPCA-1
IKK-β pIC = 7.7
F
5
0
8j
a
Values are means of at least two replicates.
Figure 4. Initial lead compounds.

D. D. Miller et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2255–2258
2257
Table 2
Enzyme inhibition by 3-substituted compounds 12–1414
Compds
R
IKK-b pIC50a
IKK-
a
pIC50a
IKK-b LE16
3
—
—
—
nPrSO
2
iPrSO
EtSO
4-Cl-PhSO
5.8
6.8
6.3
7.9
7.7
7.6
6.4
5.0
5.8
5.0
5.7
5.5
5.7
5.1
0.44
0.39
0.36
0.36
0.35
0.36
0.26
1
1
1
1
1
1
2
3
4a
4b
4c
4d
2
2
2
a
Values are means of at least two replicates.
Figure 5. Compound 3 docked into IKK-b homology model. Magenta arrows
indicate vectors for hydrogen-bonding to Asp103 and Lys106.
O
NMe2
O
OH
H
H
N
N
9
10
IKK-β pIC50 <4.8
IKK-β pIC50 <4.8
Figure 6. Modifications predicted not to bind to kinase hinge.
Figure 7. Compound 12 docked into IKK-b homology model highlighting the
putative interaction with Asp103.
targeted were Asp103 and Lys106 (Fig. 5). The first substitutents
explored at the indole 3-position were 3- and 4-piperidyl groups
(Scheme 2). These were included because of the predicted binding
mode of compound 11, a literature 2-hydroxy pyridin-2-yl IKK-b
inhibitor with a reported IC50 of 25 nM.18 In its presumed binding
mode, compound 11 accepts the hinge hydrogen-bond from Cys99
using its phenolic oxygen, while the 3-piperidine group forms a
charged hydrogen-bonding interaction with the sidechain of
Asp103.
Condensation of 3 with N-benzyl -3-piperidone or -4-piperi-
done followed by simultaneous reduction and deprotection of the
O
NH
2
O
NH2
H
N
H
N
(
a), (b)
3
1
3
Figure 8. ATP concentration dependent IKK-b inhibition by 12.
N
H
HO
N
(e)
(
c), (d)
resulting benzyltetrahydropyridine intermediate yielded the
2
H N
3
-piperidyl and 4-piperidyl products 12 and 13 respectively.
The superior potency of 12 in inhibiting IKK-b (Table 2) is attrib-
O
NH
2
O
NH2
O
NC
uted to the formation of a similar charged pair interaction between
the protonated nitrogen of the piperidine and the carboxylate side
chain of Asp103, as shown in Figure 7.
A detailed analysis of the enzyme inhibition kinetics for 12
through IC50 measurement across a range of ATP concentrations
showed an ATP concentration dependent inhibition of IKK-b, con-
sistent with the compound having an ATP-competitive mode of ac-
tion (Fig. 8).
H
N
H
N
HN
1
1
HN
N
RSO2
1
4
1
2
A further, significant increase in IKK-b inhibitory potency was
achieved though conversion of the 4-piperidyl derivative 13 into
certain sulphonamides (e.g., 14a–d). This increase was anticipated
Scheme 2. Reagents and conditions: (a) N-benzyl-4-piperidone, MeONa, MeOH,
D
D
;
;
(b) Pd(OH)
d) Pd(OH)
2
/C, H
/C, H
2
, EtOH/AcOH, rt; (c) N-benzyl-3-piperidoneꢁHCl, KOH, MeOH,
(
2
2
, EtOH, rt; (e) RSO Cl, pyridine, rt.
2

2
258
D. D. Miller et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2255–2258
Acknowledgements
The authors gratefully acknowledge Duncan Judd for technical
support in the provision of scaled-up quantities of key synthetic
intermediates, Lisa Shewchuk for obtaining the crystal structure
of 1 in CDK2, and all the members of the IKK-b team.
References and notes
1.
Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.; Sudarsanam, S. Science
002, 298, 1912.
2
2.
3.
4.
5.
6.
7.
Karin, M.; Yamamoto, Y.; Wang, Q. M. Nat. Rev. Drug Disc. 2004, 3, 17.
Pahl, H. L. Oncogene 1999, 18, 6853.
Hacker, H.; Karin, M. Sci. STKE. 2006, 357, re13.
Coish, D. G.; Wickens, P. L.; Lowinger, T. B. Expert Opin. Ther. Patents 2006, 16, 1.
Pitts, W. J.; Kempson, J. Annu. Rep. Med. Chem. 2008, 43, 155.
Bamborough, P.; Callahan, J. F.; Christopher, J. A.; Kerns, J. K.; Liddle, J.; Miller,
D. D.; Morse, M. A.; Rumsey, W. L.; Williamson, R. Curr. Topics Med. Chem. 2009,
9
, 623.
8.
Bamborough, P.; Morse, M. A.; Ray, K. P. Drug News Perspect. 2010, 23, 483.
Figure 9. Compound 14a docked into IKK-b homology model highlighting the
9. Li, J.; Peet, G. W.; Pullen, S. S.; Shembri-King, J.; Warren, T. C.; Marcu, K. B.;
putative interaction with Lys106.
Kehry, M. R.; Barton, R.; Jakes, S. J. Biol. Chem. 1998, 273, 30736.
0. Bonizzi, G.; Karin, M. Trends Immunol. 2004, 25, 280.
1
1
1. Christopher, J. A.; Avitable, B. G.; Bamborough, P.; Champigny, A. C.; Cutler, G.
J.; Dyos, S. L.; Grace, K. G.; Kerns, J. K.; Kitson, J. D.; Mellor, G. W.; Morey, J. V.;
Morse, M. A.; O’Malley, C. F.; Patel, C. B.; Probst, N.; Rumsey, W.; Smith, C. A.;
Wilson, M. J. Bioorg. Med. Chem. Lett. 2007, 17, 3972.
by docking studies which predicted that modifying the nitrogen to
an uncharged sulphonamide could permit hydrogen-bonding be-
tween an oxygen of the sulphonamide and the side-chain NH3
þ
12. Podolin, P. L.; Callahan, J. F.; Bolognese, B. J.; Li, Y. H.; Carlson, K.; Davis, T. G.;
Mellor, G. W.; Evans, C.; Roshak, A. K. J. J. Pharmacol. Exp. Ther. 2005, 312, 373.
of Lysine 106 or the backbone NH of Asp103 (Fig. 9).
1
1
3. Homology modelling, pharmacophore searching and docking were carried out
as described in Ref. 11.
Very promisingly, the enzyme inhibition of compounds 12 and
4. IKK-b kinase inhibitory activity was determined using
a time-resolved
1
4a translated into cellular efficacy in an assay measuring the com-
pounds’ ability to inhibit the LPS stimulated production of TNF- in
peripheral blood mononuclear cells (PBMC), each inhibiting with a
fluorescence resonance energy transfer (TR-FRET) assay. Recombinant GST-
tagged human IKK-b (1-737) (typically 2–5 nM final) diluted in assay buffer
a
(50 mM HEPES, 10 mM MgCl , 1 mM CHAPS pH 7.4 with 1 mM DTT and 0.01%
2
1
9
pIC50 of 6.2. Furthermore, compound 14a was found to inhibit
NF- B translocation in TNF- treated A549 cells potently, with a
pIC50 of 6.8.
w/v BSA) was added to wells containing compound or DMSO vehicle (less than
5
(
% final). The reaction was initiated by the addition of GST-I
25 nM final)/ATP (1 final), in a total volume of l. IKK-
inhibitory activity was determined similarly, using 6-His-tagged full length
IKK- . For further details see Ref.[11]. The error within both assays is estimated
j
B
a
substrate
a kinase
j
a
lM
6
l
1
9
Wider selectivity screening across an in-house panel of >45 ki-
nase assays showed an excellent profile for compounds 12 and 14a.
Both compounds had pIC50 <5.5 across the entire panel, which in-
a
as ±0.3 log units, based on the standard deviation around the mean value of an
inhibitor used as a standard compound in every assay.
15. Castro, A. C.; Dang, L. C.; Soucy, F.; Grenier, L.; Mazdiyasni, H.; Hottelet, M.;
Parent, L.; Pien, C.; Palombella, V.; Adams, J. Bioorg. Med. Chem. Lett. 2003, 13,
cluded IKK-
pIC50 = 5.9 and c-FMS pIC50 = 5.8; and 14a, BTK pIC50 = 5.8 and
Aurora A pIC50 = 5.9. In addition, 14a was screened at 10
e and TBK1, with the following exceptions: 12, BTK
2
419.
6. See discussion by Abad-Zapatero, C. Expert Opin. Drug Disc. 2007, 2, 469. Ligand
Efficiency LE = ꢀRT(ln K )/N, where N = number of heavy atoms. Here this is
approximated for relative comparision by exchanging IC50 for K
1
lM
i
20
against a panel of 105 kinases in a binding format. Where binding
was detected the K was determined. Only two kinases gave
pK > 5.5 (CSNK1 , pK = 5.8, and FLT3 pK = 5.9).
In addition to the excellent IKK-b and broader kinase profile ob-
i
.
17. Baldwin, I. R.; Bamborough, P.; Christopher, J. A.; Kerns, J. K.; Longstaff, T.;
Miller, D. D. WO 2005/067923.
d
d
e
d
d
1
8. Murata, T.; Shimada, M.; Kadono, H.; Sakakibara, S.; Yoshino, T.; Masuda, T.;
Shimazaki, M.; Shintani, T.; Fuchikami, K.; Bacon, K. B.; Ziegelbauer, K. B.;
Lowinger, T. B. Bioorg. Med. Chem. Lett. 2004, 14, 4013.
served in the series, exemplars also displayed favourable developa-
bility characteristics. For example, compound 12 had high aqueous
solubility (136 lg/ml), low protein binding (91% to Human Serum
Albumin), low in vitro clearance (human <1 ml/min/mg) and
1
2
9. Details of the PBMC and NF-jB cellular assays are described in Bamborough, P.;
Barker, M. D.; Campos, S. A,; Cousins, R. P. C.; Faulder, P.; Hobbs, H,; Holmes, D.
S.; Johnston, M. J.; Liddle, J.; Payne, J. J.; Pritchard, J. M.; Whitworth, C. WO
2008/034860.
0. The ability of compounds at 10 lM concentration to displace kinases from
encouraging oral bioavailability (22% in rat).
immobilized ATP-site probe ligands was determined as previously described,
see: 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.; Lélias, 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.
In summary, pharmacophore directed screening was used to
identify a novel IKK-b inhibitor fragment template. Homology
model-based SAR exploration led to the discovery of potent, selec-
tive inhibitors with cellular efficacy. Key exemplars such as 12 and
1
4a exceeded the minimum target profile and preliminary DMPK
data and wider profiling augured well for further optimisation of
this new class of IKK-b inhibitors, which will be described in future
publications.