Identification of a novel 3,5-disubstituted pyridine as a potent, selective, and orally active inhibitor of Akt1 kinase

Article abstract of DOI:10.1016/j.bmcl.2006.04.046

Based on lead compounds 2 and 3 a series of 3,5-disubstituted pyridines have been designed and evaluated for inhibition of AKT/PKB. Modifications at the 3 position of the pyridine ring led to a number of potent compounds with improved physical properties, resulting in the identification of 11g as a promising, orally active Akt inhibitor. The synthesis, structure-activity relationship studies, and pharmacokinetic data are presented in this paper.

Full text of DOI:10.1016/j.bmcl.2006.04.046

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3740–3744
Identification of a novel 3,5-disubstituted pyridine as a potent,
selective, and orally active inhibitor of Akt1 kinase
Sheela A. Thomas,* Tongmei Li, Keith W. Woods, Xiaohong Song, Garrick Packard,
John P. Fischer, Robert B. Diebold, Xuesong Liu, Yan Shi, Vered Klinghofer,
Eric F. Johnson, Jennifer J. Bouska, Amanda Olson, Ran Guan, Shayna R. Magnone,
Kennan Marsh, Yan Luo, Saul H. Rosenberg, Vincent L. Giranda and Qun Li
Cancer Research, Department R47S, AP10 Abbott Laboratories, 100 Abbott Park Rd, Abbott Park, IL 60064, USA
Received 29 March 2006; revised 18 April 2006; accepted 19 April 2006
Available online 5 May 2006
Abstract—Based on lead compounds 2 and 3 a series of 3,5-disubstituted pyridines have been designed and evaluated for inhibition
of AKT/PKB. Modifications at the 3 position of the pyridine ring led to a number of potent compounds with improved physical
properties, resulting in the identification of 11g as a promising, orally active Akt inhibitor. The synthesis, structure–activity relation-
ship studies, and pharmacokinetic data are presented in this paper.
Ó 2006 Elsevier Ltd. All rights reserved.
The prevalence of enhanced Akt1/protein kinase B
(PKB) activity has been observed in a large proportion
of human malignancies.1–4 The initial discovery of
Akt1, a serine-threonine kinase, as a viral oncogene5
was followed by the discovery of homology of Akt1 with
PKA and PKC kinases.6 Three cellular isoforms of Akt1
have been identified.7,8 Akt plays a pivotal role in sever-
al signal transduction pathways9–11 and is critical in the
repression of a number of apoptotic pathways.12 Over-
expression of Akt is linked to an effective survival mech-
anism for tumor cells and is associated with an increase
in tumor progression.13,14 Hence, inhibition of Akt
would be a significant tool in cancer therapy.
N
N
O
O
5
3
NH2
N
NH2
NH
NH
1
2
N
N
N
O
NH2
NH
N
3
H
Figure 1. 3,5-Disubstituted pyridines as Akt-1 inhibitors.
direction and report the synthesis and structure–activity
relationship studies of this work.
We have previously described the discovery of 3,5-disub-
stituted pyridines as novel small molecule inhibitors of
Akt1.15–17 Considerable exploration of substituents at
the 5 position of the pyridine resulted in the identifica-
tion of several potent and selective inhibitors, com-
pounds 1–3 in Figure 1. Optimization of 1 led to 2
and 3 as nanomolar inhibitors of Akt1. In order to fur-
ther investigate these lead compounds, substituents at
the 3 position of the pyridine ring were systematically
explored. In this paper, we describe our efforts in this
The route used to synthesize the 3,5-disubstituted pyri-
dines is shown in Scheme 1. The N-Boc phenols were
generated using a published procedure.18 The carboxylic
acids used were either commercially available or were
prepared by hydrolysis of the corresponding ester with
lithium hydroxide. Treatment of the acid with isobutyl
chloroformate at ꢀ10 °C, followed by a rapid reduction
with sodium borohydride, gave high yields of the alco-
hol 5. Compound 5 was treated with 3,5-bromohydroxy
pyridine under Mitsunobu conditions to generate the
3-oxo-5-bromo-pyridine compound 7. While DEAD
as well as DBAD reagents worked equally well in the
Mitsunobu reactions, the purifications of the products
Keywords: AKT/PKB; 3,5-Disubstituted pyridines; AKT inhibitors.
*
Corresponding author. Tel.: +1 847 938 2474; fax: +1 847 935
5165; e-mail: sheela.thomas@abbott.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.04.046

S. A. Thomas et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3740–3744
3741
O
with tryptophanal 13. The 30% yield obtained in this
reaction was caused by the deactivation of the aryl
amine by the pyridine ring. Introduction of the aryl
group and deprotection of the amine were carried out
as previously described.
a
R
R
OH
NHBoc
OH
NHBoc
4
5
N
N
b
+
5
Thio ether 23 was prepared as shown in Scheme 3.
Alkylation of the thiophenol 18 with dibromopyridine
17 using sodium hydride, followed by deprotection of
the methoxy benzyl group, provided compound 19.
Although Mitsunobu reactions of earlier compounds
were generally high yielding, coupling of 19 under Mits-
unobu conditions provided only a 41% yield of 21. Stille
coupling followed by removal of the Boc protecting
group provided compound 23.
Br
O
Br
OH
R
6
7
NHBoc
c
Ar-Br
ArSnMe3
d
8
N
N
e
Ar
O
Ar
O
9
R
R
NH2
NHBoc
Comparative in vitro studies of the initial compounds
prepared are summarized in Tables 1 and 2. The com-
pounds were assayed for their ability to bind to the
ATP binding site of Akt.15 At first we sought to study
the effect of replacing the indole group of the lead com-
pound 2 as shown in Table 1. Though these compounds
were less active than 2, the naphthyl analog 10d and the
indazole analog 10b were reasonably potent. A 5-fold
drop in inhibitory potency (10f vs 10i) was observed
when the chain length was shortened.
10a - 10 i Ar =
11a - 11 m Ar =
N
H3C
N
N
H
Scheme 1. Reagents and conditions: (a) 1—isobutyl chloroformate,
NMP, ꢀ10 °C; 2—NaBH4 DMF, ꢀ10 °C; (b) Ph3P, DEAD/DBAD,
THF, 0 °C to rt; (c) hexamethylditin, tetrakis(triphenylphosphine)pal-
ladium(0), toluene, 100 °C; (d) ArSnMe3, Pd2dba3, (20-dicyclohexyl-
phosphanyl-biphenyl-2-yl)-dimethyl-amine, TEA, DMF, 95 °C; (e)
TFA, CH2Cl2, rt.
A similar study of analogs of the 3-indazole compound 3
(Table 2) demonstrated the importance of the indole
group to the potency of Akt1 inhibition. Analogous to
the previous findings, all compounds prepared for this
study were less potent than the parent compound 3.
Nonetheless, many of the analogs showed nanomolar
Akt1 inhibition. Larger rings such as the naphthyl group
(11c and 11d) were tolerated in this position with the 2-
methoxynaphthyl derivative 11e showing 1.1 nM activi-
ty. The 2-methyl indole analog 11a also showed good
potency at 2.1 nM and the benzothiophene compound
11b was a fairly good Akt1 inhibitor (6.6 nM). An
were easier when the DEAD reagent was used. Aryl
stannanes 8, prepared by treatment of the bromo aryl
compound with hexamethyl ditin, were used for cou-
pling with the bromo pyridines 7 under Stille conditions
to form the N-protected compounds 9. Finally removal
of the N-Boc group under acidic conditions yielded
target compounds 10 and 11.
Synthesis of the amino compound 16 is outlined in
Scheme 2. Introduction of the amino side chain was car-
ried out by reductive amination of amino pyridine 12
HS
O
18
N
N
a
Br
Br
Br
Br
S
O
17
19
O
H
Ph
Boc
HN
N
N
13
N
N
c
b
SH
Br
S
a
Br
NH2
Br
NH
20
21
12
14
NH
HN
BOC
HN
Boc
N
d
N
b
Ar
S
Ar
NH
NH
HN
R
HN
22: R= BOC
23: R= H
R
e
15: R= BOC
16: R= H
c
Scheme 3. Reagents and conditions: (a) NaH, DMF, 48 h; (b) TFA,
m-cresol, reflux; (c) 5, Ph3P, DBAD, THF, 0 °C to rt; (d) ArSnMe3,
Pd2dba3, (20-dicyclohexylphosphanyl-biphenyl-2-yl)-dimethyl-amine,
TEA, DMF, 95 °C overnight; (e) TFA, CH2Cl2, rt.
Scheme 2. Reagents and conditions: (a) NaBH3CN, Ti(iPrO)4, EtOH,
rt; (b) ArSnMe3, Pd2dba3, (20-dicyclohexylphosphanyl-biphenyl-2-yl)-
dimethyl-amine, TEA, DMF, 95 °C; (c) TFA, CH2Cl2, rt.

3742
S. A. Thomas et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3740–3744
Table 1. Akt1 inhibition of 5-isoquinolinyl pyridines
Table 2. Akt1 inhibition of 5-indazolyl pyridines
N
N
R
R
N
N
N
H
R
Compound
Akt1 inhibition Ki (nM)
R
Compound
Akt1 inhibition Ki (nM)
O
2
2
O
3
0.16
NH2
NH2
NH
N
NH2
NH
NH
S
O
O
10a
10b
10c
10d
10e
10f
10g
10h
10i
54.6
6.8
O
11a
11b
11c
11d
2.1
6.6
3.2
8.3
NH2
O
O
NH2
N
NH
S
NH2
NH2
O
O
63.7
3.5
NH2
NH2
O
NH2
O
85
NH2
N
O
O
11e
11f
11g
11h
11i
11j
1.1
185
NH
2O
O
O
O
89.6
1525
672
465
NH2
NH2
NH2
NH2
N
N
O
O
O
11
NH2
NH2
NH2
3282
368
N
O
NH2
Values were measured against Akt1 with ATP concentration of 10 lM.
233
O
NH2
aromatic ring is required in this position as seen by the
loss in potency in going from the phenyl analog 11g
(11 nM) to the cyclohexyl analog (368 nM). The quino-
line compounds exhibited diminished activity in both
series (10e and 11f) and the 4-pyridyl replacements
resulted in a complete loss of activity (10g and 11h).
Values were measured against Akt1 with ATP concentration of 10 lM.
displayed excellent oral bioavailability (67%). While
the naphthyl derivatives, in particular 11c, did exhibit
improved PK over the parent 3, other more potent com-
pounds demonstrated either lower or no oral bioavail-
ability. Interestingly, the isoquinoline analog of 11g,
compound 10f, was not orally bioavailable.
The effect of a decrease or an increase in chain length
was demonstrated by 11j–11m in comparison to 11g
(Table 3) and was seen to result in a significant drop
in Akt inhibition. Replacement of the ether linkage of
compound 11g with a nitrogen linkage (16) resulted in
a more potent compound, while replacement of the ether
linkage of 3 with sulfur (23) proved deleterious to
activity.
Antiproliferative activity of the Akt inhibitors was eval-
uated in a MTT assay15 and is illustrated in Table 5. Cell
growth inhibition determined in MiaPaCa-2 human
pancreatic cancer cells indicated reasonable cellular
activity in most cases.
While enzymatic potencies of compounds 2 and 3 were
extremely high, we have reported earlier that they exhib-
ited extremely poor aqueous solubility and this was det-
rimental to the potential development of these
compounds. The mouse pharmacokinetic profile of the
more potent currently described compounds was evalu-
ated and is shown in Table 4. In this study, 11g
The excellent pharmacokinetic profile of 11g prompted
us to carry out a more detailed study of the pharmaco-
kinetics of this compound in other species as shown in
Table 6. The results clearly indicate significant oral bio-
availability with a good clearance and half-life across
species.

S. A. Thomas et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3740–3744
Table 7. Selectivity profile of 11g for different kinases
3743
Table 3. 5-Indazolyl pyridine analogs
N
Kinase
AKT-1 inhibition Ki (nM)
AKT-1
PKA
11
16
R
N
N
H
PKCy
PKCd
PDK1
CDK2
ERK2
GSK3b
MAPK-AP2
CK2
1200
360
R
Compound
Akt1 inhibition Ki (nM)
>20,000
46
260
O
11k
289
NH2
110
1100
5400
>3700
13,000
5800
>2200
2600
580
O
O
11l
11m
16
197
70.6
8.3
NH2
KDR
SRC
cKIT
FLT1
NH2
CHK1
RSK2
N
H
NH2
S
23
244
626
(24) as well. We used the route depicted in Scheme 1 for
the preparation of 24. As seen in Table 3, compound 24
exhibited greatly diminished Akt1 inhibition, reiterating
the role played by the chirality of the amine toward
Akt1 inhibition.
NH2
NH2
NH
O
24
Values were measured against Akt1 with ATP concentration of 10 lM.
The selectivity profile of 11g was assessed against a large
panel of kinases and is shown in Table 7. Selectivity is
always a challenge for kinase inhibitors and the results
we obtained were as we had expected. While 11g exhib-
ited considerable selectivity against less closely related
kinases, the high degree of homology between PKA
and Akt1 was reflected in the lack of selectivity against
PKA.
Table 4. PK profile of select Akt1 inhibitors in mouse (10 mg/kg po)
Compound
Akt1 inhibition
IC50 (nM)
PK PO AUC
F (%)
2
2
0
0
0
3
0.16
3.5
89.6
2.1
6.6
3.2
1.1
11
0
10d
10f
11a
11b
11c
11e
11g
16
0.073
0
3.9
0
In summary, we have demonstrated that replacement of
the indole moiety in 3,5-disubstituted pyridine analogs
can lead to compounds retaining Akt1 inhibitory activ-
ity. We selected several compounds for further evalua-
0
0
0.42
0.524
0.394
2.01
0
4.2
27.6
17
67
0
tion and found
a number of compounds with
8.3
improved pharmacokinetic properties. We further iden-
tified a compound, 11g, as a novel, potent, selective, and
orally bioavailable inhibitor of Akt1 kinase.
Table 5. Cellular activity in MiaPaCa-2
Compound 10d 10f 11a 11b 11c 11e 11g 16
EC50 (lM) 0.4 0.1 18 1.0 0.2 0.4 1.9 0.3 0.4 0.2
2
3
Acknowledgment
We thank Dr. Thomas Penning for helpful discussions
and for critical reading of the manuscript.
Table 6. PK profile of 11g across different species
Species Dose (mg/kg) T1/2 Vb Cl PO AUC F (%)
5.0 2.0
1.6 0.94
0.9 1.83
References and notes
Mouse
Rat
10
5
1.0
3.6
3.1
7.6
8.4
3.9
67
30
63
13
1. Bellacosa, A.; Testa, J. R.; Moore, R.; Larue, L. Cancer
Biol. Ther. 2004, 3, 268.
2. Mitsades, C. S.; Mitasades, N.; Koutsilieris, M. Curr.
Cancer Drug Targets 2004, 4, 235.
3. Hanada, M.; Feng, J.; Hemmings, B. A. Biochem. Biophys.
Acta 2004, 1697, 3.
Dog
2.5
Monkey 2.5
NC* NC* <0.49 0.68
* Not calculated.
4. Vivanco, I.; Sawyers, C. L. Nat. Rev. Cancer 2002, 2, 489.
5. Staal, S. P. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 5034.
6. Jones, P. F.; Jakubiwicz, T.; Pitossi, F. J.; Maurer, F.;
Hemmings, B. A. Proc. Natl. Acad. Sci. U.S.A. 1991, 88,
4171.
Since, as described in our earlier work, the S-enantiomer
was responsible for Akt1 potency, we had prepared only
the S-isomers of our analogs. However, the promising
profile of 11g prompted us to prepare the R-enantiomer

3744
S. A. Thomas et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3740–3744
7. Cheng, A. Q.; Godwin, A. K.; Bellacosa, A. Proc. Natl.
Acad. Sci. U.S.A. 1992, 89, 9267.
8. Nakatani, K.; Sakaue, H.; Thompson, D. A.; Weigel, R.
J.; Roth, R. A. Biochem. Biophys. Res. Commun. 1999,
257, 906.
9. Paez, J.; Sellers, W. R. Cancer Treat. Res. 2003, 115, 145.
10. Kumar, R.; Hung, M.-C. Cancer Res. 2005, 65, 2511.
11. Kandel, E. S.; Hay, N. Exp. Cell Res. 1999, 253, 210.
12. Downward, J. Semin. Cell Dev. Biol. 2004, 15, 177.
13. Testa, J.; Bellacosa, A. Proc. Natl. Acad. Sci. U.S.A. 2001,
98, 10983.
14. Page, C.; Lin, H. J.; Jin, Y.; Castle, V. P.; Nunez, G.;
Huang, M.; Lin, J. Anticancer Res. 2000, 20, 407.
15. Luo, Y.; Shoemaker, A.; Liu, X.; Woods, K.; Thomas, S.;
De Jong, R.; Han, E.; Li, T.; Stoll, V.; Powlas, J.;
Oleksijew, A.; Mitten, M.; Shi, Y.; Guan, R.; McGonigal,
T.; Klinghofer, V.; Johnson, V.; Leverson, J.; Bouska, J.;
Mamo, M.; Smith, R.; Gramling-Evans, E.; Zinker, B.;
Mika, A.; Nguyen, P.; Oltersdorf, T.; Rosenberg, S.; Li,
Q.; Giranda, V. Mol. Cancer Ther. 2005, 4, 977.
16. Li, Q.; Woods, K. W.; Thomas, S.; Zhu, G.-D.;
Packard, G.; Fisher, J.; Li, T.; Gong, J.; Dinges, J.;
Song, X.; Abrams, J.; Shi, Y.; Liu, X.; Klinghofer, V.;
De Jong, R.; Oltersdorf, T.; Stoll, V.; Jakob, C.;
Rosenberg, S.; Giranda, V. Bioorg. Med. Chem. Lett.
2006, 16, 2000.
17. Woods, K. W.; Fisher, J. P.; Claiborne A.; Li, T.; Thomas,
S. A.; Zhu, G-D.; Diebold R. B.; Liu. X.; Shi, Y.;
Klinghofer, V.; Han, E. K.; Guan, R.; Magnone, S. R.;
Johnson, E. F.; Bouska, J. J.; Olson, A.; De Jong, R.;
Oltersdorf, T.; Luo, Y.; Giranda, V.; Li, Q. J. Med.
Chem., submitted for publication.
18. Rodriguez, M.; Llinares, M.; Doulut, S.; Heitz, A.;
Martinez, J. Tetrahedron Lett. 1991, 32, 923.