Synthesis, initial SAR and biological evaluation of 1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-4-amine derived inhibitors of IκB kinase

Article abstract of DOI:10.1016/j.bmcl.2009.03.159

A new series of tricyclic-based inhibitors of IKK have been derived from an earlier lead compound. The synthesis and structure-activity relationships (SAR) are described. Compound 4k inhibited TNF production in rats stimulated with LPS.

Full text of DOI:10.1016/j.bmcl.2009.03.159

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2646–2649
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis, initial SAR and biological evaluation of 1,6-dihydroimidazo[4,5-
d]pyrrolo[2,3-b]pyridin-4-amine derived inhibitors of IjB kinase
*
James Kempson , Junqing Guo, Jagabandhu Das, Robert V. Moquin, Steven H. Spergel, Scott H. Watterson,
Charles M. Langevine, Alaric J. Dyckman, Mark Pattoli, James R. Burke, XiaoXia Yang, Kathleen M. Gillooly,
Kim W. McIntyre, Laishun Chen, John H. Dodd, Murray McKinnon, Joel C. Barrish, William J. Pitts
Bristol-Myers Squibb Research and Development, Princeton, NJ 08543-4000, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A new series of tricyclic-based inhibitors of IKK have been derived from an earlier lead compound. The
synthesis and structure–activity relationships (SAR) are described. Compound 4k inhibited TNF produc-
tion in rats stimulated with LPS.
Received 9 March 2009
Revised 27 March 2009
Accepted 30 March 2009
Available online 5 April 2009
Published by Elsevier Ltd.
Keywords:
IKK
TNF
SAR
LPS
The nuclear transcription factor NF-
j
B has a crucial role in the
most effective anti-inflammatory agents available, is mediated
primarily through the inhibition of NF- B mediated transcrip-
tion. This suggests that an IKK2 inhibitor would provide an
opportunity to develop novel therapeutics which has the poten-
tial for glucocorticoid-like activity but without their use-limiting
toxicities.5
pathogenesis of several human disorders, particularly those with
an inflammatory component.1 In unstimulated cells, NF-
B is
sequestered in the cytoplasm as an inactive complex with I
inhibitory protein.2 In the case of I
, the most extensively stud-
j
j
jB
jBa
ied member of this protein family, the stimulation of pro-inflam-
matory receptors such as the TNF receptor activates the so-called
The discovery of small molecule IKK inhibitors as orally active
therapeutic agents for the treatment of inflammatory diseases
has been the goal of numerous research groups, and many classes
of IKK inhibitors have been reported.6 We recently reported the
identification of several tricyclic based inhibitors of IKK2 (Fig. 1),
1a, 1b and 1c.6h
‘canonical’ pathway of NF-
jB activation, in which a multi-subunit
IjB kinase (IKK) complex catalyses the phosphorylation of IjBa
at
Ser32 and Ser36. This phosphorylation event is essential for signal-
ing the subsequent ubiquitination and protosomal degradation of
IjBa, thus leaving the NF-jB free to translocate to the nucleus
and activate pro-inflammatory gene transcription.3
The IKK complex is comprised of two catalytic subunits, IKK1
and IKK2. Although both of these subunits can catalyse the phos-
N
N
N
N
N
O
N
S
phorylation of I
subunit plays a major role in NF-
pro-inflammatory stimuli such as tumor necrosis factor TNF-
IL1-b, IL-6, IL-17 and IL-23. Since IKK2 is vital for translating
these pro-inflammatory stimuli into the activation of NF- B, an
jB protein, evidence has shown that the IKK2
N
N
H
jB activation arising from
N
N
H
1a
1b
a,
IKK2 IC50 11nM
IKK1 IC50 270nM
PBMC IC50 390nM
IKK2 IC50 6nM
IKK1 IC50 3500nM
PBMC IC50 310nM
j
N
N
inhibitor of IKK2 could be effective in the treatment of autoim-
mune and inflammatory disorders such as rheumatoid arthritis,
inflammatory bowel disease, lupus and multiple sclerosis.4 Inter-
estingly, the activity of glucocorticoids, which are among the
N
N
N
N
H
1c
IKK2 IC50 520nM
IKK1 IC50 4280nM
PBMC IC50 446nM
* Corresponding author. Tel.: +1 609 252 4539.
E-mail address: james.kempson@bms.com (J. Kempson).
Figure 1. Tricyclic inhibitors of IKK2.
0960-894X/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2009.03.159

J. Kempson et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2646–2649
2647
The previously reported chemistry effort6h was greatly facili-
tated by using a common intermediate, 2, which allowed our lab
to build upon the SAR observations established across these, and
other, earlier chemotypes. As part of a multi-chemotype strategy
to isosteric inhibitors of IKK kinase, we sought to extend our earlier
work and proposed an ‘azaindole’ core 4. A synthetic route was de-
vised that relied upon a 5-endo-dig cyclisation7 from an appropri-
ately substituted acetylene precursor 3 (Fig. 2). Since we had an
abundant supply of 2 we devised a route to 3 which could take
advantage of this advanced intermediate.
The synthetic pathway utilized in the preparation of the azain-
dole system is outlined in Scheme 1. The amine functionality in the
common intermediate, 2, was removed by diazotization followed
by reduction with hypophosphorous acid to give the compound
5, in quantitative yield. Regiospecific displacement of one of the
chloro groups with methylamine followed by Boc-protection led
to 7 in 68% yield over two steps. All attempts to displace the
remaining chloropyridine in 7 with a variety of nitrogen nucleo-
philes failed, even under forcing microwave conditions. After some
experimentation it was discovered that Buchwald amination con-
ditions using benzophenone imine8 as the nitrogen source in com-
bination with Pd2(dba)3, Xanthphos and cesium carbonate in DMF
led to an excellent 95% yield of 8. Mild hydrolysis of the imine in 8
was achieved with 1 M aqueous HCl in THF to reveal the 2-amino
pyridine 9. This was then reacted with exactly 1 equiv of NIS in
acetonitrile to give the late stage iodo-intermediate 10 in 80%
yield. Sonagashira coupling of 10 with phenyl acetylene in the
presence of a catalytic amount of PdCl2(PPh3)2 and CuI using
DIPA as base provided 11a (R = H) in 75% yield. Formation of the
‘azaindole’ ring was effected with potassium tert-butoxide7 to pro-
vide the desired tricyclic system, 12a (R = H). Final deprotection
using 4 M HCl in dioxane produced the desired compound 4a.
The analogs reported in Table 1 were synthesized in a similar
manner using either commercially available acetylenes or coupling
partners synthesised according to a general two-step procedure as
illustrated in Scheme 2 for 3-ethynyl-benzonitrile.
N
N
H2N
Cl
In the case of 15, further elaboration was carried out as high-
lighted in Scheme 3 to afford final compound 18. The nitrile in
17 was reduced under hydrogen using Ra-Ni as catalyst. This pro-
vided the benzylamine 19, which then formed the basis for subse-
1a, 1b, 1c
N
2
Cl
Common Intermediate
R
TMS
N
N
R
I
N
R
N
PdCl2(PPh3)2, CuI
NC
DMF, rt (95%)
NC
N
H
H2N
N
N
H
N
N
H
13
3
4
14 R = TMS
15 R = H
cat. KOH
in MeOH
(95%)
Figure 2. Proposed azaindole core from common intermediate, 2.
Scheme 2. Synthesis of acetylene coupling partners.
N
N
N
N
H2N
Cl
H
MeNH2, EtOH
reflux (86%)
NaNO2, H3PO4
100 ºC (100%)
N
N
Cl
15
Cl
N
Cl
NC
KOtBu
NC
N
10
2
5
PdCl2(PPh3)2, CuI
DMF, rt (95%)
DMA, 80 ºC
(80%)
H2N
N
N
Boc
NH
N
N
16
NaHMDS, Boc2O
-78 ºC (79%)
N
N
Ph
Ph
N
H
H
N
Pd2(dba)3, Xantphos
Cs2CO3, dioxane (95%)
N
N
Cl
N
Cl
N
N
H
N
H2, Raney-Ni
(75%)
Boc
7
6
N
H
N
N
Boc
N
H
N
N
R
NH2
NC
N
N
N
H
N
N
H
HCl, THF
(90%)
NIS, CH3CN
(80%)
17 R = Boc
18 R = H
19 R = Boc
20 R = H
HCl in dioxane
HCl in dioxane
N
N
Boc
H2N
N
N
Boc
Ph
Ph
9
8
N
N
R
R
HCl in dioxane
(80-95%)
EDC, HOBt
RCO2H
N
N
N
H
N
N
Boc
N
N
KOtBu
I
NH
O
R
PdCl2(PPh3)2, CuI
DMF, rt (25-86%)
DMA, 80 ºC
(50-85%)
(30-78%)
H2N
N
N
Boc
21g-l
H2N
N
N
Boc
11
10
N
N
N
R
N
N
R
HCl in dioxane
(80-95%)
N
N
H
N
N
H
N
H
N
N
H
N
H
NH
O
R
N
N
Boc
4g-l
4
12
Scheme 1. General azaindole synthetic route.
Scheme 3. Azaindole scaffold amenable to further derivatization.

2648
J. Kempson et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2646–2649
Table 1
quent coupling with a variety of carboxylic acids under standard
SAR with respect to IKK and PBMC inhibition
carbodiimde mediated conditions. Final deprotection of the Boc
protecting group with HCl in dioxane provided the compounds
4g–l.
N
R
N
N
H
Compounds were evaluated for their ability to inhibit IKK1 or
IKK2 catalyzed phosphorylation of GST-I
a peripheral blood mononuclear cell assay (PBMC) was used to
N
N
H
jBa
fusion protein9 and
Compd
R
IKK2 IC50 (nM)
40
IKK1 IC50 (nM)
1670
PBMC IC50 (nM)
800
measure the ability of compounds to inhibit LPS-induced TNF
a
production in human primary cells.10 The structure–activity rela-
tionships for the inhibition of IKK and PBMC activity are summa-
rized in Table 1.
4a
4b
44
NTa
931
A direct comparison between the previous tricyclic systems
(1a–c) and the first azaindole compound, 4a, shows nearly a seven-
fold loss in IKK2 activity against the most potent thiazole com-
pound 1b, with ꢀ2-fold loss in PBMC activity compared to all
three previous tricycles. Fluorine substitution (4b, 4c, 4d) gave
essentially equipotent inhibitors to 4a for ortho- and para-substitu-
tion. The meta-substituted analog, 4c, however displayed almost a
6-fold increase in IKK2 and PBMC potency over 4a. Meta methoxy
compound 4c possessed similar potency. The meta-nitrile com-
pound 18 proved to be optimal with potent IKK2 inhibition and
excellent translation into activity in cells. Unfortunately the phys-
ical characteristics of this compound precluded its evaluation
F
4c
4d
4e
7
43
8
NTa
153
538
320
F
F
1730
NTa
MeO
NC
18
20
7
NTa
NTa
86
in vivo (e.g., poor solubility <5 lg/mL). An attempt to improve
the physicochemical properties began with the benzylamine ana-
log, 20, displaying potent IKK2 inhibition and modest PBMC activ-
ity. Derivatization of 20 provided a range of amide analogs
containing alkyl (4f), alcohol (4g), heterocyclic (4h), amine (4i),
amide (4j) and ether (4k) functionalities. Despite the wide toler-
ance of functionality by the enzyme, cell potency for 4i and 4j
was significantly reduced compared to all other analogs studied.
4k was the most potent IKK2 analog studied in the amide series.
Based on its good IKK1 selectivity and acceptable PBMC activity,
together with good physical characteristics imparted by the side
chain (aqueous solubility at pH 1.0 = 1.823 mg/mL), it was decided
to evaluate this compound further.
11
340
NH2
4f
14
25
633
NTa
372
478
NH
O
O
4g
NH
In a rat pharmacokinetic study (Table 2), 4k was administered
orally at 10 mg/kg and at 2 mg/kg intravenously. The compound
had a favorable half-life (t1/2) of 4.8 h, and a mean residence time
(MRT) of 2.1 h. The measured oral bioavailability (Fpo) in this
experiment was acceptable at 31%. The steady-state volume of dis-
tribution (Vss) was low as was the systemic clearance (Cl).
The in vivo efficacy of 4k was next assessed in an acute rat mod-
HO
4h
4i
39
79
150
NTa
NTa
NH
S
O
O
N
N
el of TNFa
inhibition.11 Rats were dosed orally with 4k at 3 mg/kg,
10 mg/kg and 30 mg/kg, 4 h prior to LPS administration. The TNF
a
levels were then measured 90 min later.
As shown in Figure 3, oral dosing of 4k inhibited TNFa produc-
tion in a dose-dependent manner. Statistically significant inhibi-
2000
NH
N
tion of TNF production (85%) was observed at an oral dose of
30 mg/kg. Given the rat whole blood IC50 value of 8
the exposure observed at the 30 mg/kg dose (17 M) is consistent
with this significant inhibition of TNF. In contrast, the 10 mg/kg
lM for 4k,
l
dose was not effective, consistent with an exposure (3
lM) below
NH
4j
22
NTa
1000
12
O
the rat whole blood IC50
.
O
H2N
Table 2
Rat PKb data for 4k following iv and po dosing
Dosea
(mg/kg)
AUCtot
(nM h)
t1/2
(h)
MRT
(h)
Cl
Vss
(L/kg)
Fpo
(%)
(mL/min/kg)
NH
4k
12
2620
370
O
iv, 2
po, 10
46923
71747
4.8
11.0
2.1
14.7
1.8
0.2
31
OMe
a
Vehicle: iv, propylene glycol/water (80/20); po 100% propylene glycol.
n = 3 animals.
b
a
NT = not tested.

J. Kempson et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2646–2649
2649
Marinier, A.; Dodier, M.; Martel, A.; Nirschl, D.; Van Kirk, K.; Burke, J. R.; Pattoli,
M. A.; Gillooly, K.; McIntyre, K. W.; Chen, L.; Yang, Z.; Marathe, P. H.; Wang-
Iverson, D.; Dodd, J. H.; McKinnon, M.; Barrish, J. C.; Pitts, W. J. J. Med. Chem.,
2009, 52, 1994.
7. Rodriguez, A. L.; Koradin, C.; Knochel, P. Angew. Chem., Int. Ed. 2000, 14, 2488.
8. Wolfe, J. P.; Åhman, J.; Sadighi, J. P.; Singer, R. A.; Buchwald, S. L. Tetrahedron
Lett. 1997, 38, 6367.
9. Assays measuring the enzyme-catalyzed phosphorylation of GST-I
performed by adding enzyme (IKK-2, typically to a final concentration of 3
mL) at room temperature to solutions of 50 g/mL GST-I and 20 M ATP in
jB
a
were
l
g/
l
j
Ba
l
25 mM TrisÁHCl, pH 7.5, containing 7.5 mM MgCl2, 34 mM sodium phosphate,
3 mM NaCl, 0.6 mM potassium phosphate, 1 mM KCl, 1 mM dithiothreitol, 5%
(w/v) glycerol, and 470
reactions were stopped by the addition of EDTA to 33 mM.
phosphorylation was quantitated by competition for binding to an anti-
phospho-I antibody (SantaCruz Biotechnology #sc-8404) with fluorescein-
lg/mL bovine serum albumin. After 60 min, the kinase
IjBa
jBa
labeled phospho-peptide ([FL]-Asp-Asp-Arg-His-Asp-[p]Ser-Gly-Leu-Asp-Ser-
Met-Lys-NH2) as measured using fluorescence polarization.
10. Human PBMCs were isolated from from whole blood collected from healthy
donors. Blood was diluted into RPMI 1640 (Life Technologies) containing
Figure 3. Inhibition of LPS-induced serum TNFa in rats (n = 8/group, mean SD).
2.5 mM EDTA (Life Technologies), 10 lg/mL polymyxin (Sigma), and then
underlaid with ficoll (Accurate Scientific Co.) and centrifuged at 600g for
25 min. The interface was collected, and cells were washed twice and
resuspended in RPMI, 10% FBS. Cells are then distributed (200 mL/well) into
96-well tissue culture treated plates (Falcon) at 1 Â 106 cells/mL in RPMI, 10%
FBS. Test compounds were added to appropriate wells and incubated with cells
for 30 min. Cells were then stimulated by the addition of lipopolysaccharide
(LPS, BioWhittaker), with a final concentration of 25 ng/mL, and incubated for
6 h at 37 °C, 5% CO2. The cell supernatants were removed and assayed for TNF-
In summary, we have reported the generation of a novel tricy-
clic scaffold utilizing the same common intermediate from previ-
ous efforts. This chemotype displays efficacy in an acute model
of inflammation and current efforts to improve the whole blood
potency of this series continues and will be reported at a later time.
a
by ELISA (R&D Systems).
References and notes
11. LPS-induced Serum TNF-
a
in Rats: Male Lewis rats (175–200 g, Harlan) were
dosed by oral gavage with vehicle (50% PEG400, 50% 0.02 N HCl) or 4k in
vehicle. Four hours later, rats were injected intraperitoneally with 1 mL per rat
of 100 lg lipopolysaccharide (LPS, Escherichia coli strain 0111:B4, Sigma-
1. (a) Viemann, D. Blood 2004, 103, 3365; (b) Neurath, M. F. Gut 1998, 43, 856; (c)
Cho, M. L. J. Immunol. 2006, 176, 5652; (d) Zhang, Z. Int. J. Mol. Med. 2005, 15,
79; (e) He, K. L.; Ting, A. T. Eur. J. Immunol. 2003, 33, 1917; (f) Snapper, C. M. J.
Immunol. 1996, 156, 183; (g) Ruocco, M. G. J. Exp. Med. 2005, 201, 1677; (h)
Pattoli, M. A. J. Pharmacol. Exp. Ther. 2005, 315, 382.
2. Whiteside, S. T.; Israel, A. Sem. Cancer Biol. 1997, 8, 75.
3. Ghosh, S.; Karin, M. Cell 2002, 109, S81.
4. Strnad, J.; Burke, J. R. Trends Pharmacol. Sci. 2007, 28, 142.
5. Saklatvala, J. Arthritis Res. 2002, 4, 146.
6. (a) Podolin, P. L. J. Pharmacol. Exp. Ther. 2005, 312, 373; (b) Burke, J. R. J. Biol.
Chem. 2003, 278, 1450; (c) Ziegelbauer, K. Br. J. Pharmacol. 2005, 145, 178; (d)
Wen, D. J. Pharmacol. Exp. Ther. 2006, 317, 989; (e) Belema, M.; Bunker, A.;
Nguyen, V. N.; Beaulieu, F.; Ouellet, C.; Qiu, Y.; Zhang, Y.; Martel, A.; Burke, J. R.;
McIntyre, K. W.; Pattoli, M. A.; Daloisio, C.; Gillooly, K. M.; Clarke, W. J.; Brassil,
P. J.; Zusi, F. C.; Vyas, D. M. Bioorg. Med. Chem. Lett. 2007, 17, 4284; (f) Qiu, Y.;
Belema, M.; Yang, X.; Zusi, F. C.; Pitts, W. J. PCT Int. Appl. WO 2004075846,
2004.; (g) Beaulieu, F.; Ouellet, C.; Ruediger, E. H.; Belema, M.; Qiu, Y.; Yang, X.;
Banville, J.; Burke, J. R.; Gregor, K. R.; MacMaster, J. F.; Martel, A.; McIntyre, K.
W.; Pattoli, M. A.; Zusi, F. C.; Vyas, D. Bioorg. Med. Chem. Lett. 2007, 17, 1233; (h)
Kempson, J.; Spergel, S. H.; Guo, J.; Quesnelle, C.; Gill, P.; Belanger, D.; Dyckman,
A. J.; Li, T.; Watterson, S. H.; Langevine, C. M.; Das, J.; Moquin, R. V.; Furch, J. A.;
Aldrich, St. Louis, MO) suspended in phosphate buffered saline. Ninety minutes
after LPS challenge, blood was collected from the retro-orbital sinus under
isoflurane anesthesia. Serum was separated from clotted samples by
centrifugation (5 min, 5000gÂ, room temperature) and analyzed for the
levels of TNF-a by ELISA according to the manufacturer’s instructions
(BioSource, Camarillo, CA). Results are shown as mean SD of n = 8 rats per
treatment group. Unless otherwise noted, statistical significance was
determined using a one-way analysis of variance (ANOVA) with a Dunnett’s
post-test. Values of p < 0.05 were considered significant.
12. For rat whole blood experiments, blood was collected by cardiac puncture in
collection tubes containing ACD-a as an anticoagulant. 300 lL of pooled whole
blood was pipetted into deep-well 96 well assay plates (Costar), and inhibitor
in phosphate buffered saline containing 16% (v/v) DMSO was then added to
give a final DMSO concentration of 0.2% (v/v). After incubating 45 min at 37 °C,
lipopolysaccahride (LPS, S.Typhosa, Sigma) was added to a final concentration
of 5 lg /mL. After a 4 h incubation at 37 °C, the plates were centrifuged for
10 min at 14,000 rpm and plasma removed for TNF
(Biosource).
a measurement by ELISA