The design and synthesis of potent and cell-active allosteric dual Akt 1 and 2 inhibitors devoid of hERG activity

Article abstract of DOI:10.1016/j.bmcl.2008.05.084

This letter details the attenuation of hERG in a class of Akt inhibitors through heteroatom insertions into aromatic rings. The development of a cell-active dual Akt 1 and 2 inhibitors devoid of hERG activity is discussed using structure-activity relationships.

Full text of DOI:10.1016/j.bmcl.2008.05.084

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4191–4194
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The design and synthesis of potent and cell-active allosteric dual Akt 1 and 2
inhibitors devoid of hERG activity
b
a
a
a
a
b
Tony Siu a, , Yiwei Li , Johnny Nagasawa , Jun Liang , Lida Tehrani , Peter Chua , Raymond E. Jones ,
*
Deborah Defeo-Jones b, Stanley F. Barnett b, Ronald G. Robinson b
a Department of Medicinal Chemistry, Merck Research Laboratories, Merck & Co., 3535 General Atomics Court, San Diego, CA 92129, USA
b Department of Cancer Research, Merck Research Laboratories, Merck & Co., PO Box 4, West Point, PA 19486, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
This letter details the attenuation of hERG in a class of Akt inhibitors through heteroatom insertions into
aromatic rings. The development of a cell-active dual Akt 1 and 2 inhibitors devoid of hERG activity is
discussed using structure–activity relationships.
Received 3 March 2008
Revised 18 May 2008
Accepted 19 May 2008
Available online 24 May 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
hERG
Allosteric
Akt
Kinase
Akt is a serine/threonine kinase belonging to the AGC family of
kinases. Activation of Akt is responsible for multiple cellular events
including cell survival, energy metabolism, proliferation, and
growth. In an earlier letter, we identified potent dual Akt 1 and 2
inhibitors by incorporating solubilizing amines.1a However, these
inhibitors as well as other inhibitors discovered in our program
represented by 11b and 21a (Fig. 1) possess off-target binding affin-
ity to the Ikr potassium channel hERG (human Ether-a-go-go-Re-
lated Gene) with IC50s in the micromolar range. Affinity for the
hERG channel has the potential to alter cardiac repolarization
through prolongation of the QT interval,2 and alterations of the
QT interval could potentially trigger torsades de pointes, a life-
threatening ventricular tachyarrhythmia. Because of the potential
for cardiac death associated with QT prolongation, optimization
of the hERG binding window is a goal in the development of a drug.
This letter outlines our strategy to identify a cell-based potent dual
Akt 1 and 2 inhibitors devoid of hERG activity.
ability of our inhibitors to penetrate cells. Our strategy was to
incorporate a variety of polar group modifications to the B ring of
1 to address hERG binding and to evaluate their effects on cellular
potency.
The analogs for our strategy commenced with hydrazide forma-
tion 3 using carbonyl diimidazole and hydrazine illustrated in
Scheme 1. Heating hydrazide 4 with a variety of nitriles together
in the presence of sodium methoxide provided access to an assort-
ment of polar aromatic end groups as represented by 5. These com-
pounds were attached to subunit pyridone 61b by reductive
amination to afford 7.
As shown in Table 1, changes in polarity to the aromatic B ring
attenuated hERG-binding affinity. For example, significant
improvements were observed with pyrimidine 8, pyridone 9, and
amino pyridine 10. These compounds all displayed binding affinity
of IC50 = 10 l
M or greater to hERG.5 In addition, these substitutions
were not detrimental to the overall intrinsic or cell-based potency
Earlier work at Merck Research Laboratories has shown that
minor heteroatom additions to aromatic rings can attenuate hERG
binding.3 Presumably these modifications work by decreasing the
for Akt 1 and 2. Other heteroaromatic derivatives such as pyrazine
11 (hERG IC50 = 8.5 lM) achieved modest gains in attenuating
hERG binding, while no improvements were obtained with
2-methoxy pyridine 12 (hERG IC50 = 1220 nM). Gratifyingly, in all
instances the intrinsic and cell-based potencies were within 3-fold
of 1 while modifying hERG binding.
lipophilicity of the aromatic ring, thus destabilizing
p stacking
interactions with the hERG-binding environment.4 However, it
was not clear whether these changes in polarity would affect the
With hERG binding attentuated through heteroatom additions
to the aromatic B ring, we turned our attention to merging this
key finding with our strategy to enhance cell-based potency. Our
conception was that the potency optimized piperazines discovered
in the previous communication1a would confer cellular potency
* Corresponding author at present address: Department of Drug Optimization,
Merck Research Laboratories, Merck & Co., 33 Avenue Louis Pasteur, Boston, MA
02115, USA.
E-mail address: tony_siu@merck.com (T. Siu).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.05.084

4192
T. Siu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4191–4194
N
N
N
H
N
N
B
N
H
N
B
N
N
A
N
A
N
N
N
N
HN
N
N
N
O
1
2
Akt 1/2 IC50 = 3.5/42 nM
Akt 1/2 IC50 = 4/4 nM
Cell Akt 1/2 IC50 = 16/266 nM
hERG IC50 = 2500 nM
Cell Akt 1/2 IC50 = 10/39 nM
hERG IC50 = 2100 nM
Figure 1. Merck allosteric Akt inhibitors.
O
NHNH2
O
OH
X
X
X
X
X
H
N
a
b,c
HN
X
N
N
X
X
N
N
NC
Boc
Boc
X
5
3
4
X
X
N
X
X
X
H
O
N
N
N
d
HN
HN
N
N
O
O
6
7
Scheme 1. Synthesis of pyridone analogs. Reagents and conditions. (a) Hydrazine, carbonyl diimidazole, CH2Cl2; (b) nitrile, 25% sodium methoxide in MeOH, 2-ethoxyet-
hanol, 100 °C; (c) TFA, CH2Cl2; (d) 6, sodium triacetoxyborohydride, DMF.
Table 1
whereas the polar aromatic end groups would impart favorable
hERG-binding properties on the pyridopyrimidine core (Fig. 2).
Toward that end, our strategy for optimizing both cell-based
potency and reducing hERG affinity was to systematically synthe-
size 12 combinations of the most potent piperazine subunits with
the least potent hERG-binding aromatic B rings in a matrix strat-
egy. This 4 Â 3 matrix is detailed in Table 2. The piperazines chosen
for this optimization were methyl piperazine, hydroxylethyl piper-
azine, and dimethyl amino piperazine, whereas the aromatic B ring
subunits chosen were pyrimidine, amino pyridine, pyridone, and
pyridine N-oxide. The analogs were synthesized using the same
route as previously detailed.1a
Consistent with observations in Table 1, incorporation of the
polar heteroaromatic end groups in general attenuated hERG bind-
ing. Of the 12 compounds synthesized, compound 13 proved opti-
mal in terms of cell-based potency (cell Akt 1/2 IC50 = 25/40 nM)
and hERG binding (hERG IC50 > 10,000 nM). The hydroxy ethylpip-
erazine series (17–20) maintained low hERG-binding affinity;
however, the cellular potency diminished slightly against Akt 2.
On the other hand, the dimethyl amino piperazine series (21–24)
SAR of hERG binding with polar end groups
N
H
N
N
R
N
HN
N
O
7
R
Enzyme IC50 (nM)
Cell IC50 (nM)
hERG IC50 (nM)
Akt 1
Akt 2
Akt 1
Akt 2
N
4
26
18
224
69
2500
1
N
7
1
4
4
3
44
8
23
62
26
44
40
>10,000
>10,000
10,000
8428
N
8
NH
284
165
405
159
O
9
N
33
21
16
NH2
10
N
R
X
X
N
N
X
X
X
H
N
N
N
N
N
11
N
N
N
Off Target Reducing
N
Potency Enhancing
1220
OMe
12
Figure 2. Strategy for combining potency and reducing hERG affinity.

T. Siu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4191–4194
4193
Table 2
Matrix SAR with solubilizing amines and polar end groups
N
H
R1
N
N
R2
N
N
N
N
Cell Akt 1/2 nM
(hERG nM)
R2
N
N
NH
N
N
O
NH2
O
71/188
NA
20/84
6100
14
88/185
9800
16
25/40
N
(>10000)
N
N
13
15
R1
HO
N
37/80
38/180
7400
127/284
6800
67/146
>10000
(>10000)
17
18
19
20
N
N
33/68
7900
37/140
3500
69/98
800
35/130
4800
21
N
22
23
24
also had slightly diminished cell-based potency, but still possessed
low micromolar hERG binding affinity. The loss in cell-based po-
tency in Akt 2 for the ethylhydroxyl and dimethyl amino pipera-
zines might be due to large polarity increase in the molecule
when coupled to the aromatic B ring end groups. The methyl piper-
azine in combination with the pyrimidine end group appears to
contain the appropriate amount of polarity to decrease hERG bind-
ing, but do not affect the cellular activity of the molecule. This
strategy proved useful in optimizing the overall polarity for opti-
mal cell-based potency with decreased hERG binding affinity.
Based on the success of inserting heteroatoms into the B aro-
matic ring to reduce hERG activity, we also explored whether het-
eroatom insertion into the A ring can bring the same success.
Toward that end, our strategy was to construct a pyridine–pyra-
zine core as opposed to a pyrimidine–pyridine core (Fig. 3).
The synthesis of Akt inhibitors with a pyridine–pyrazine core
commenced with an ammonia displacement of chloride at the 2-
position of 2,6-dichloro-3-nitro-pyridine to provide 26. Reduction
of the nitro moiety with tin(II) chloride in ethyl acetate gener-
ated the diamine 27.6 Condensation of diamine 27 with bromo-
benzoin 28 afforded a 1:1 mixture of inseparable isomers 29.
With 29 in hand, the piperidine subunit 30,1b was attached to
the core by displacement with K2CO3 in dimethylformamide to
afford 31. Heating 31 with excess methylpiperazine in dioxane
followed by HPLC separation of the regioisomers provided 32
and 33 (Scheme 2).
a high level of cell-based potency against both Akt 1 and 2 (18/
92 nM) as well as attenuated hERG binding (9100 nM). In contrast,
the opposite regioisomer 33 suffered a loss in activity against Akt 1
and 2 (1408/827 nM) and did not attenuate hERG binding when
compared to 2.
Having established the low hERG-binding affinity for this scaf-
fold, we turned to our matrix strategy to further optimize potency
and hERG parameters using the same piperazines and similar het-
eroaromatic end groups as illustrated in Table 2 (Table 3). From
this optimization strategy, compound 36 was identified as pos-
sessing favorable hERG binding (IC50 hERG >10,000 nM) along
with high cellular potency (cell Akt 1 and 2 IC50: 13/48 nM). This
compound showed comparable activities to 13. The combination
of the N-(2-hydroxyethyl)piperazinyl and the 2-pyridinyl end
groups proved to be optimal. Interestingly, N-(2-hydroxyethyl)
piperazines 37 and 38 also had favorable hERG binding affinities,
but had only modest cellular Akt 2 activities. In contrast to the
pyrimidine-pyridine scaffold in Table 2, the polar pyrimidine
and 2-aminopyridine end groups rendered the molecules too po-
lar when coupled to the pyridine–pyrazine core. This drastic in-
crease in polarity was detrimental to the cellular activity.
Presumably, the pyridine–pyrazine core coupled with a pipera-
zine was polar enough to affect the hERG binding without the
added polarity of Table 1 end groups. When either the methyl
piperazine or the N,N-dimethyl amino piperazine was substituted
on the A ring, either a lack of cellular potency or increased hERG
binding was observed. Compounds 32 and 39 showed good cellu-
lar potency against Akt 1/2, but the attenuation of hERG binding
was modest when compared to 2. Similar to the results in Table
2, these data indicate that an overall balanced molecule polarity
was necessary for the optimal cell-based potency and hERG
attenuation.
From our previous work, it was anticipated that regioisomer 32
would possess the more desirable cell-based potency.1a As antici-
pated, the pyridine–pyrazine core from regioisomer 32 possesses
In summary, a strategy to reduce hERG activity for dual Akt 1
and 2 allosteric inhibitors was outlined. Increasing the overall
polarity of the molecule through small heteroatom insertions into
aromatic rings proved successful. By coupling this key finding for
modulating hERG activity with our previous discovery of using
water solublizing piperazines for potency enhancement, inhibitors
13 and 36 were identified as low-nanomolar cell-based inhibitors
N
A
N
RN
N
N
A
RN
N
N
Figure 3. Alternative core heterocycle.

4194
T. Siu et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4191–4194
Cl
N
NH2
NH2
Cl
N
NH2
NO2
Cl
N
Cl
c
b
a
NO2
27
26
25
Br
N
H
N
Cl
N
N
N
Cl
N
N
N
d
e
N
N
N
regioisomer
regioisomer
31
29
N
H
N
N
H
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
32
33
Cell Akt 1/2 IC50 = 1408/827 nM
hERG IC50 = 2143 nM
Cell Akt 1/2 IC50 = 18/92 nM
hERG IC50 = 9100 nM
Br
HN
H
O
O
N
N
N
N
28
30
Scheme 2. Synthesis of pyrido-pyrazine core. Reagents and conditions. (a) Ammonium hydroxide, EtOH, 100 °C; (b) Tin (II) Chloride, EtOAc; (c) 28, AcOH, 100 °C, (d)
potassium carbonate, 30, DMF; (e) methyl piperazine, dioxane, 100 °C.
Acknowledgments
Table 3
SAR on pyridine–pyrazine analogs
The authors thank Drs. Christopher Cox, Christopher Dinsmore,
and Philip Sanderson for careful reading of this manuscript.
N
H
N
R1
N
N
N
R2
N
N
References and notes
1. (a) Siu, T.; Liang, J.; Arruda, J.; Li, Y.; Jones, R. E.; Defeo-Jones, D.; Barnett, S. F.;
Robinson, R. G. Bioorg. Med. Chem. Lett. 2008, 18, 4186; (b) Bilodeau, M. T.;
Balitza, A. E.; Hoffman, J. M.; Manley, P. J.; Barnett, S. F.; Defeo-Jones, D.; Jones, R.
E.; Robinson, R. G.; Smith, A. M.; Huber, H. E.; Hartman, G. D. Bioorg. Med. Chem.
Lett. 2008, 18, 3178.
Cell Akt 1/2 nM
(hERG nM)
R2
N
N
N
2. (a) Fermini, B.; Fossa, A. A. Nat. Rev. Drug Discov. 2003, 2, 439; (b) Roden, D. M. N.
Engl. J. Med. 2004, 350, 1013.
N
NH2
3. (a) Bell, I. M.; Gallicchio, S. N.; Abrams, M.; Beshore, D. C.; Buser, C. A.;
Culberson, J. C.; Davide, J.; Ellis-Hutchings, M.; Fernandes, C.; Gibbs, J. B.;
Graham, S. L.; Hartman, G. D.; Heimbrook, D. C.; Homnick, C. F.; Huff, J. R.;
Kassahun, K.; Koblan, K. S.; Kohl, N. E.; Lobell, R. B.; Lynch, J. J., Jr.; Miller, P.
A.; Omer, C. A.; Rodrigues, A. D.; Walsh, E. S.; Williams, T. M. J. Med. Chem.
2001, 44, 2933; For an excellent review of hERG Optimization in Medicinal
Chemistry, see: (b) Jamieson, C.; Moir, E. M.; Rankovic, Z.; Wishart, G. J. Med.
Chem. 2006, 49, 5029.
N
70/225
(4700)
35
52/192
(8300)
34
18/92
N
N
(9100)
R1
32
HO
N
13/48
72/190
115/230
(>10000)
(>10000)
(>10000)
4. (a) Mitcheson, J. S.; Chen, J.; Lin, M.; Culberson, C.; Sanguinetti, M. C. Proc. Natl.
Acad. Sci. U.S.A. 2000, 97, 12329; (b) Fernandez, D.; Ghanta, A.; Kauffman, G. W.;
Sanguinetti, M. C. J. Biol. Chem. 2004, 279, 10120.
36
37
38
N
N
5. The hERG IC50 values were acquired by radioligand-binding competition
experiments using membrane preparations from human embryonic kidney
cells that express hERG. For assay details, see Ref.3 and references therein. IC50
values are reported as the averages of at least two independent determinations;
standard deviations are within 25–50% of IC50 values.
104/116
10/43
NA
N
(>10000)
(6000)
39
40
6. Oguchi, M.; Wada, K.; Honma, H.; Tanaka, A.; Kaneko, T.; Sakakibara, S.; Ohsumi,
J.; Serizawa, N.; Fuijiwara, T.; Horikoshi, H.; Fujita, T. J. Med. Chem. 2000, 43,
3052.
against Akt 1 and 2 with greater than 10,000 nM binding against
hERG.