Synthesis and structure based optimization of novel Akt inhibitors

Article abstract of DOI:10.1016/j.bmcl.2008.04.034

Based on a high throughput screening hit, pyrrolopyrimidine inhibitors of the Akt kinase are explored. X-ray co-crystal structures of two lead series results in the understanding of key binding interactions, the design of new lead series, and enhanced pot

Full text of DOI:10.1016/j.bmcl.2008.04.034

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 3359–3363
Synthesis and structure based optimization of novel Akt inhibitors
Blaise Lippa,* Gonghua Pan, Matthew Corbett, Chao Li, Goss S. Kauffman,
Jayvardhan Pandit, Shaughnessy Robinson, Liuqing Wei, Ekaterina Kozina,
Eric S. Marr, Gary Borzillo, Elisabeth Knauth, Elsa G. Barbacci-Tobin,
Patrick Vincent, Merin Troutman, Deborah Baker, Francis Rajamohan,
Shefali Kakar, Tracey Clark and Joel Morris*
Pfizer, Inc., PGRD Groton, 558 Eastern Point Road, Groton, CT 06340, USA
Received 6 March 2008; revised 8 April 2008; accepted 10 April 2008
Available online 15 April 2008
Abstract—Based on a high throughput screening hit, pyrrolopyrimidine inhibitors of the Akt kinase are explored. X-ray co-crystal
structures of two lead series results in the understanding of key binding interactions, the design of new lead series, and enhanced
potency. The syntheses of these series and their biological activities are described. Spiroindoline 13j is found to have an Akt1 kinase
IC50 of 2.4 0.6 nM, Akt cell potency of 50 19 nM, and provides 68% inhibition of tumor growth in a mouse xenograft model
(50 mg/kg, qd, po).
Ó 2008 Elsevier Ltd. All rights reserved.
The serine/threonine kinase Akt (PKB) is a key signaling
component in the IGF-1R/PI3K/PTEN cell survival
pathway.1–4 It was initially identified as a fusion protein
in an oncogenic retrovirus (AKR mouse Thymoma),
and was subsequently shown to have oncogenic poten-
tial in cell transformation assays. Three Akt genes exist
in humans: AKT1, AKT2, and AKT3.
designed to bind in the enzyme active site in competition
with ATP.13–19
At Pfizer, a high throughput screen revealed that pyrrol-
opyrimidine 1 was a 290 nM inhibitor of Akt1 in an in
vitro kinase assay (Fig. 1).20 To further explore the
SAR of this lead, multiple analogs were synthesized
(Scheme 1). Condensation of 4-hydrazinoquinazolines
(2) with orthoformates followed by ring opening with
aqueous KOH provided anilinotriazole 3.21 The aniline
was then added to a chloropyrrolopyrimidine (4)22 in
1,2-dichloroethane resulting in the completed analogs
(5).
Multiple observations point to Akt as an important can-
cer drug discovery target,5 including: (i) the tumor sup-
pressor PTEN, a negative regulator of Akt kinase
activity, is mutated or deleted at high frequency in solid
human cancers and several cancer susceptibility syn-
dromes6; (ii) Akt is activated via growth factor receptors
and/or ligands that are up-regulated in a wide variety of
solid human tumors, and (iii) AKT gene amplification
has been reported in several cancer lines.7,8
The resulting compounds were assayed for the inhibition
of Akt1 (Table 1).20 Methyl substitution of the pyrrole
Based on the strong rationale for inhibiting Akt as a
therapy for cancer, there have been multiple efforts to
discover small molecule antagonists.9,10 These have
included both allosteric inhibitors,11,12 and inhibitors
HN
N
N
NH
N
N
H
N
1
Akt1 kinase IC50 = 210 +/- 66 nM
Keywords: Akt; Pyrrolopyrimidine; Spiroindoline; Imidazopiperidine.
*
Figure 1. Structure and Akt1 kinase potency of high throughput
screen hit.
Corresponding authors. Tel.: +1 781 860 8402 (B.L.); e-mail:
blaise.lippa@cubist.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.04.034

3360
B. Lippa et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3359–3363
R2
NHNH2
1) R2C(OEt)3
HN
N
N
N
R1
N
R1
2) aq. KOH
35-66%
NH2
3
2
R2
R3
Cl
N
HN
N
4
N
R1
N
R3
N
NH
5
R4
4
N
1,2-dichloroethane
34-83%
N
N
R4
5
Scheme 1. The synthesis of anilinotriazole pyrrolopyrimidine analogs.
Table 1. Akt1 kinase potency of anilinotriazole analogs 1 and 5a–f
R2
HN
N
4
5
N
R1
R3
NH
N
N
N
5
R4
Compound R1
R2
R3
R4
a
Akt1 kinase IC50 (nM)
Figure 2. X-ray co-crystal structure of triazole 5d bound to Akt1.
1
H
H
H
H
H
H
H
H
H 210 66
Me >10,000
5a
5b
5c
5d
5e
5f
H
tency. In designing new analogs of this type, it was also
deemed important to preserve the intermolecular hydro-
gen bonds that the triazole makes with bound waters.
With these design criteria in mind, imidazopiperidine
analogs were targeted.
H
Me
Cl
H
H
H
H
H
H
425 112
318 86
151 32
171 51
>10,000
H
Me
4-Cl Me
5-Cl Me
H
H
a Values are means of at least two experiments.
Scheme 2 illustrates the synthesis of imidazopiperidines
8a and 8b. Histamine (6) is condensed with paraformal-
dehyde in refluxing water to form imidazopiperidine
7.24 The amine is then added to chloropyrrolopyrimi-
dine (4) in low yield (unoptimized) to obtain the final
analogs. Imidazopiperidine analog 8a inhibited Akt1
with similar potency as triazole 1 (294 80 nM). How-
ever, in this series chlorine substitution at C5 of the pyr-
rolopyrimidine provided a 7-fold boost in potency (8b:
42 30 nM), representing a key potency lead in antago-
nizing Akt1. This substituent effect is divergent from the
nitrogen (5a) resulted in a complete loss of activity, sug-
gesting that this NH and the adjacent pyrimidine nitro-
gen may be forming a key donor–acceptor hydrogen
bonding interaction with the hinge region of the protein,
common among kinase inhibitors. Substitution at the
C5 position on the pyrrolopyrimidine nucleus with
either methyl or chloro was tolerated (5b, 5c), as was
methyl substitution at R2 of triazole (5d). SAR on the
linking phenyl ring showed more divergent SAR. While
a Cl group was tolerated at the 4 position (5e), a Cl
group at the 5 position resulted in >10 lM inhibition
of Akt1 (5f).
N
N
Subsequently, an X-ray co-crystal structure of analog 5d
bound to the kinase domain of Akt1 was generated (Fig.
2).23 This structure confirmed that the pyrrolopyrimi-
dine was in fact acting as a hydrogen bond donor–accep-
tor in binding to the hinge region of Akt1. The crystal
structure further reveals that the triazole is forming
two hydrogen bonds to adjacent waters and is pi stack-
ing with phenylalanine 161. Importantly, the solved
structure also showed a tight intramolecular hydrogen
bond between the aniline NH and one of the triazole
nitrogens. This observation suggested that additional
rigidity could be introduced into the molecule by
replacement of this hydrogen bond with a covalent
bond, resulting in a new ring, and possibly improved po-
NH
(CH2O)n
NH
H2O, reflux
90%
N
H
NH2
6
7
N
Cl
N
R
NH
R
N
N
N
N
H
N
4
N
H
R = H: K2CO3/ H2O, reflux; 31%
R = Cl: NaHCO3/ DMSO, 80 C; 20%
º
8a: R = H
8b: R = Cl
Scheme 2. Synthesis of imidazopiperidine analogs.

B. Lippa et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3359–3363
3361
Cbz
Cbz
N
R1
N
i.) cat. TFA/ Tol., ACN
R1
+
NH2
ii.) NaBH4/ MeOH
40-90%
N
H
9
N
CHO
10
H
R2
N
11
R1
R3
Cl
N
N
R3
N
N
N
N
H
4
N
H3PO4, KH2PO4
DMSO, 80
62-80%
H
ºC
12: R2 = Cbz
13: R2 = H
TFA, 70
62-93%
ºC
Scheme 3. Synthesis of spiroindoline analogs.
spiroindoline 11. Addition of this intermediate to chlo-
ropyrrolopyrimidine 4 initially proved challenging.
Although a melt at high temperatures with these re-
agents provided product, it was often low yielding and
not reproducible. Subsequently, it was found that acid
catalysis in DMSO gave improved yields, presumably
through the protonation of the pyrimidine resulting in
an increase in the electrophilicity of the heterocycle.
However, particularly with cyanide substituents, acid
catalysis along with the HCl generated in the reaction
as a by-product led to decomposition and low yields.
An optimum procedure was developed using phosphoric
acid to catalyze the reaction in the presence of KH2PO4
as a buffer to consume the HCl by-product. This proce-
dure resulted in yields of 62–80% across a wide range of
substrates.
Figure 3. X-ray co-crystal structure of imidazopiperidine 8b bound to
Akt1.
triazole series SAR where a C5 substituent provided no
boost in potency (1 vs 5b, 5c).
To better understand the binding of imidazopiperidines
to Akt, an X-ray co-crystal structure of 8b was solved
(Fig. 3).23 While the pyrrolopyrimidine portion of 8b
bound nearly identically with that of 5d (Fig. 2), the re-
gion around the imidazopiperidine showed large
changes. Most notably, the protonated imidazole nitro-
gens are still involved in two hydrogen bonds, but in-
stead of binding to bound waters, they are binding to
glutamate 234 and aspartate 292 in the vicinity of where
the ribose of ATP would normally occupy. In addition,
phenylalanine 161, which pi stacked with the triazole of
5d, is now swung outward resulting in a significant con-
formational change in this portion of the protein.
The unsubstituted spiroindoline 13a proved to be a very
potent Akt inhibitor, with 5.3 0.9 nM kinase po-
tency,20 and 96 9 nM cell potency26 (Table 2). This
lead was tolerant to substituents at R1 on the phenyl
ring (13b, 13c), similar to the triazole series (5e). Sepa-
rately, while a methyl R2 substituent on the piperidine
of the spiroindoline was relatively well tolerated (13d),
functionality that reduced the basicity of the piperidine
nitrogen (13e, 13f) was significantly deleterious to activ-
ity. Substitution at C5 on the pyrrolopyrimidine (R3) re-
vealed that chloro (13g) and methyl (13h) substitution
was tolerated. This SAR is similar to the triazole series
(5b, 5c), but different from the imidazopiperidine series
(8b) where chlorine substitution led to a large increase
in potency. In the spiroindoline series, further increases
in bulk at R3 to Bn (13i) caused a large drop in potency,
whereas cyanide 13j was well tolerated with a kinase
IC50 of 2.4 0.6 nM and cell potency of 50 19 nM.27
Compounds in all series showed little kinase selectivity
over the closely related PKA. For instance, 13j inhibits
PKA with an IC50 of 3.6 2.2 nM, under identical
conditions.
From an SAR perspective, the enhanced potency of this
cyclized amine moiety suggested the need to more
broadly explore analogs incorporating this functional-
ity. Toward this goal, a diverse set of amines were ap-
pended to chloropyrrolopyrimidines 4, and from this
effort spiroindolines (13, Scheme 3) were identified as
potent inhibitors. Analogs of this type contain a cyclic
amine fused to a phenyl group. The cyclic amine of 13
is expected to overlap well with the piperidine of imidaz-
opiperidine 8, whereas the fused phenyl ring is expected
to overlap well with the phenyl of triazole leads (1, 5).
The distal spiro-piperidine amine of 13 is expected to
bind in the region of the imidazole/triazole, and thus
form hydrogen bonds with the polar residues in this por-
tion of the Akt1 binding pocket.
Given the impressive activity profile and ligand effi-
ciency28 of 13j, it was further studied to determine its
in vivo anti-tumor and pharmacokinetic properties.
Mice xenografted with a Rat-1a tumor line activated
with human myristilated Akt1 (Myr-Akt1) were given
50 mg/kg doses of 13j for 10 days orally. This treatment
Compounds of this type were synthesized through the
addition of optionally substituted hydrazines (9) with
the protected aldehyde 10 (Scheme 3).25 Cyclization fol-
lowed by reduction with sodium borohydride resulted in

3362
B. Lippa et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3359–3363
Table 2. Akt kinase and cell potency of spiroindolines 13a– j
R2
N
R1
R3
N
N
N
N
13
H
a
Akt1 kinase IC50 (nM)
a
Compound
R1
R2
R3
Akt cell IC50 (nM)
96
51 33
13a
13b
13c
13d
13e
13f
13g
13h
13i
H
Cl
F
H
H
5.3 0.9
4.8 1.2
3.1 1.9
9
H
H
H
H
75
6
H
F
Me
COMe
SO2Me
H
H
12.7 11.0
2790 318
3190 346
5.8 4.9
136
NT
NT
3
H
F
H
H
H
H
H
Cl
Me
Bn
64 49
122 19
NT
H
4.8 3.9
1330 572
2.4 0.6
H
13j
H
CN
50 19
a Values are means of at least two experiments; NT, not tested.
7. Carpten, J. D.; Faber, A. L.; Horn, C.; Donoho, G. P.;
Briggs, S. L.; Robbins, C. M.; Hostetter, G.; Boguslawski,
S.; Moses, T. Y.; Savage1, S.; Uhlik, M.; Lin, A.; Du, J.;
Qian, Y.-W.; Zeckner, D. J.; Tucker-Kellogg, G.; Touch-
man, J.; Patel, K.; Mousses, S.; Bittner, M.; Schevitz, R.;
Lai, M.-H. T.; Blanchard, K. L.; Thomas, J. E. Nature
2007, 448, 439.
8. Pedrero, J. M. G.; Carracedo, D. G.; Pinto, C. M.;
Zapatero, A. H.; Rodrigo, J. P.; Nieto, C. S.; Gonzalez,
M. V. Int. J. Cancer 2005, 114, 242.
9. Li, Q. Expert Opin. Ther. Patents 2007, 17, 1077.
10. Lindsley, C. W.; Barnett, S. F.; Yaroschak, M.; Bilodeau,
M. T.; Layton, M. E. Curr. Topics Med. Chem. 2007, 7,
1349.
11. Lindsley, C. W.; Zhao, Z.; Leister, W. H.; Robinson, R.
G.; Barnett, S. F.; Defeo-Jones, D.; Jones, R. E.;
Hartman, G. D.; Huff, J. R.; Huberb, H. E.; Duggan,
M. E. Bioorg. Med. Chem. Lett. 2005, 15, 761.
resulted in tumor growth inhibition of 68% versus con-
trol, with plasma Cmax, Cave and total daily AUC con-
centrations of 705 ng/mL, 402 ng/mL, and 3216 ng h/
mL, respectively.
In conclusion, the use of structure based drug design
aided in the optimization of a high throughput screening
hit, resulting in the design of several new series and a
ꢀ100-fold increase in Akt1 inhibition. Compound 13j
was further shown to inhibit Akt in cells, and to slow
the growth of tumors in vivo. Kinase selectivity remains
a key concern. Further results in optimizing kinase selec-
tivity, Akt potency, and drug-like properties will be pub-
lished in due course.
Acknowledgments
12. Kozikowski, A. P.; Sun, H.; Brognard, J.; Dennis, P. A.
J. Am. Chem. Soc. 2003, 125, 1144.
Michael Luzzio, Kevin Freeman-Cook, Chiliu Chen,
Christopher Autry, Catherine Hulford, Kendra Nelson,
Gary Martinelli, Donn G. Wishka, Matthew Marx,
Yong Lu, Martin Wythes, Chakrapani Subramanyam,
William Hungerford, Jinshan Chen, Shang-Poa Chang,
Joel Arcari, Tricia Kwan, Carl Thompson, and Aaron
Kanter provided additional chemistry support. Addi-
tional biology support was received from Kevin Cole-
man, Kristina Rafidi, Kou Kou, Jing Lin, Erik
Soderstrom, and Lili Yao.
13. Zhu, G.-D.; Gandhi, V. B.; Gong, J.; Thomas, S.; Woods,
K. W.; Song, X.; Li, T.; Diebold, R. B.; Luo, Y.; Liu, X.;
Guan, R.; Klinghofer, V.; Johnson, E. F.; Bouska, J.;
Olsen, A.; Marsh, K. C.; Stoll, V. S.; Mamo, M.;
Polakowski, J.; Campbell, T. J.; Martin, R. L.; Gintant,
G. A.; Penning, T. D.; Li, Q.; Rosenberg, S. H.; Giranda,
V. L. J. Med. Chem. 2007, 50, 2990.
14. Luo, Y.; Shoemaker, A. R.; Liu, X.; Woods, K. W.;
Thomas, S. A.; De Jong, R.; Han, E. K.; Li, T.; Stoll,
V. S.; Powlas, J. A.; Oleksijew, A.; Mitten, M. J.; Shi,
Y.; Guan, R.; McGonigal, T. P.; Klinghofer, V.;
Johnson, E. F.; Leverson, J. D.; Bouska, J. J.; Mamo,
M.; Smith, R. A.; Gramling-Evans, E. E.; Zinker, B.
A.; Mika, A. K.; Nguyen, P. T.; Oltersdorf, T.;
Rosenberg, S. H.; Li, Q.; Giranda, V. L. Mol. Cancer
Ther. 2005, 4, 977.
15. Lin, X.; Murray, J. M.; Rico, A. C.; Wang, M. X.; Chu,
D. T.; Zhou, Y.; Del Rosario, M.; Kaufman, S.; Ma, S.;
Fang, E.; Crawford, K.; Jefferson, A. B. Bioorg. Med.
Chem. Lett. 2006, 16, 4163.
16. Breitenlechner, C. B.; Wegge, T.; Berillon, L.; Graul, K.;
Marzenell, K.; Friebe, W.-G.; Thomas, U.; Schumacher,
R.; Huber, R.; Engh, R. A.; Masjost, B. J. Med. Chem.
2005, 48, 163.
References and notes
1. Bellacosa, A.; Franke, T. F.; Gonzalez-Portal, M. E.
Oncogene 1993, 8, 745.
2. Fayard, E.; Tintignac, L.; Baudry, A.; Hemmings, B. J.
Cell Sci. 2005, 24, 7455.
3. Altomare, D. A.; Testa, J. R. Oncogene 2005, 24, 7455.
4. Borzillo, G. V. Drug Discov. Today: Ther. Strateg. 2005, 2,
331.
5. LoPiccolo, J.; Granville, C. A.; Gills, J. J.; Dennis, P. A.
Anti-Cancer Drugs 2007, 18, 861.
6. Samuels, Y.; Ericson, K. Curr. Opin. Oncol. 2006, 18, 77.
17. Lippa, B. Expert Opin. Ther. Patents 2007, 17, 577.

B. Lippa et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3359–3363
3363
18. Donald, A.; McHardy, T.; Rowlands, M. G.; Hunter, L.-
J. K.; Davies, T. G.; Berdini, V.; Boyle, R. G.; Aherne, W.;
Garrett, M. D.; Collins, I. J. Med. Chem. 2007, 50, 2289.
19. Saxty, G.; Woodhead, S. J.; Berdini, V.; Davies, T. G.;
Verdonk, M. L.; Wyatt, P. G.; Boyle, R. G.; Barford, D.;
Downham, R.; Garrett, M. D.; Carr, R. A. J. Med. Chem.
2007, 50, 2293.
20. Akt1 kinase assay: A fluorescence polarization IMAP type
assay is used. 15 lL of diluted test compound in DMSO is
mixed with 60 lL of reaction buffer (10 mM Tris–HCl, pH
7.5, 10 mM MgCl2, 0.1 mM EGTA, 0.01% Triton X-100,
1 mM DTT). About 5 lL of the compound/buffer mixture,
10 lL of a solution containing 4 lM ATP, and 40 nM
fluorescent-labeled Crosstide (Tamara-labeled GRPRT
SSFAEG peptide), and 5 lL of Akt1 protein (lacking the
pleckstrin homology (PH) domain, containing an Asp at
position 473, and pre-phosphorylated at Thr 308) in
reaction buffer are combined. After a 90 min incubation,
IMAP beads are added and plates are read (lamp filter:
544 nm, emission filter: 615 nm). The same procedure can
be applied to full length Akt1 to provide similar results. For
example, compound 1 inhibits full length Akt1 with an IC50
of 223 110 nM. As expected, no compounds tested have
shown Akt isozyme selectivity (Akt1, 2 or 3).
22. Gerster, J. F.; Hinshaw, B. C.; Robins, R. K.; Townsend,
L. B. J. Heterocycl. Chem. 1969, 6, 207.
23. Pandit, J. WO 2005113762A1, 2005; Atomic coordinates
and structure factors of the protein structures in this paper
have been deposited in the RCSB Protein Data Bank, with
accession numbers 3CQU and 3CQW.
24. Habermehl, G. G.; Ecsy, W. Heterocycles 1976, 5, 127.
25. Maligres, P. E.; Houpis, I.; Rossen, K.; Molina, A.; Sager,
J.; Upadhyay, V.; Wells, K. M.; Reamer, R. A.; Lynch, J.
E.; Askin, D.; Volante, R. P.; Reider, P. J.; Houghton, P.
Tetrahedron 1997, 53, 10983.
26. Akt cell assay: The phosphorylation status of an Akt
substrate, GSK-3a, in the U87 glioblastoma line is
measured. Cells were incubated with compound (7 con-
centrations plus vehicle) for 2 h in DMEM medium lacking
fetal calf serum (FCS). After cell lysis, p-GSK-3a was
captured on 96-well plates coated with a phosphospecific
antibody to GSK-3a/b (#9331, Cell Signaling Technol-
ogy). The reporter antibody bound total GSK-3a (#SC-
5264, Santa Cruz), and was measured in an ELISA format.
27. Unfortunately, it was not possible to obtain high-quality
crystal structures on any member of the spiroindoline lead
series (13).
28. Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discov.
Today 2004, 9, 430.
21. Potts, K. T.; Brugel, E. G. J. Org. Chem. 1970, 35, 3448.