Discovery and optimization of new benzimidazole- and benzoxazole-pyrimidone selective PI3Kβ inhibitors for the treatment of phosphatase and TENsin homologue (PTEN)-deficient cancers

Article abstract of DOI:10.1021/jm300241b

Most of the phosphoinositide-3 kinase (PI3K) kinase inhibitors currently in clinical trials for cancer treatment exhibit pan PI3K isoform profiles. Single PI3K isoforms differentially control tumorigenesis, and PI3Kβ has emerged as the isoform involved in the tumorigenicity of PTEN-deficient tumors. Herein we describe the discovery and optimization of a new series of benzimidazole- and benzoxazole-pyrimidones as small molecular mass PI3Kβ-selective inhibitors. Starting with compound 5 obtained from a one-pot reaction via a novel intermediate 1, medicinal chemistry optimization led to the discovery of compound 8, which showed a significant activity and selectivity for PI3Kβ and adequate in vitro pharmacokinetic properties. The X-ray costructure of compound 8 in PI3Kδ showed key interactions and structural features supporting the observed PI3Kβ isoform selectivity. Compound 8 achieved sustained target modulation and tumor growth delay at well tolerated doses when administered orally to SCID mice implanted with PTEN-deficient human tumor xenografts.

Full text of DOI:10.1021/jm300241b

Article
pubs.acs.org/jmc
Discovery and Optimization of New Benzimidazole- and
Benzoxazole-Pyrimidone Selective PI3Kβ Inhibitors for the Treatment
of Phosphatase and TENsin homologue (PTEN)-Deficient Cancers
Victor Certal,† Frank Halley,*,† Angela Virone-Oddos,† Cecile Delorme,† Andreas Karlsson,‡ Alexey Rak,‡
́
Fabienne Thompson,† Bruno Filoche-Romme,† Youssef El-Ahmad,† Jean-Christophe Carry,†
́
Pierre-Yves Abecassis,§ Pascale Lejeune,† Loic Vincent,† Helene Bonnevaux,† Jean-Paul Nicolas,†
́
̀
Thomas Bertrand,‡ Jean-Pierre Marquette,‡ Nadine Michot,∥ Tsiala Benard,⊥ Peter Below,#
Isabelle Vade,† Fabienne Chatreaux,† Gilles Lebourg,† Fabienne Pilorge,† Odile Angouillant-Boniface,†
Audrey Louboutin,† Christoph Lengauer,† and Laurent Schio†
†Oncology Drug Discovery, ‡Structural Biology and Molecular Modeling, §Drug Safety, ∥Protein Production, ⊥Formulation, #Parallel
Synthesis, Sanofi Research & Development, 13 quai Jules Guesde, 94403 Vitry-sur-Seine, France
ABSTRACT: Most of the phosphoinositide-3 kinase (PI3K)
kinase inhibitors currently in clinical trials for cancer treatment
exhibit pan PI3K isoform profiles. Single PI3K isoforms
differentially control tumorigenesis, and PI3Kβ has emerged
as the isoform involved in the tumorigenicity of PTEN-deficient
tumors. Herein we describe the discovery and optimization of a
new series of benzimidazole- and benzoxazole-pyrimidones as
small molecular mass PI3Kβ-selective inhibitors. Starting with
compound 5 obtained from a one-pot reaction via a novel
intermediate 1, medicinal chemistry optimization led to the
discovery of compound 8, which showed a significant activity
and selectivity for PI3Kβ and adequate in vitro pharmacokinetic
properties. The X-ray costructure of compound 8 in PI3Kδ
showed key interactions and structural features supporting the observed PI3Kβ isoform selectivity. Compound 8 achieved
sustained target modulation and tumor growth delay at well tolerated doses when administered orally to SCID mice implanted
with PTEN-deficient human tumor xenografts.
subunit represent the only enzymes clearly implicated in human
cancer.1,2 Most of the PI3K inhibitors currently in clinical
development are pan-PI3K inhibiting all four class I PI3K
isoforms. However, recent reports support the clinical develop-
ment of isoform-specific inhibitors.3,4 Indeed, first clinical trials
with pan-PI3K inhibitors performed on nonstratified cancer
patient cohorts appear to be less successful than expected, with
no clear genetic determinant of sensitivity observed yet. On the
basis of preclinical studies, while PI3Kα-specific inhibition is
predicted to block the growth of tumors harboring PIK3CA
mutations, PTEN-deficient tumors have been shown to depend
on PI3Kβ activity. Indeed, it has been reported that conditional
knockout of PIK3CB, and not PIK3CA, leads to tumor growth
inhibition in PTEN-deficient genetic context, as well as to
prevent prostate tumor formation induced by PTEN loss with
concomitant diminution of AKT phosphorylation.5,6 Selective
compounds against PI3Kα or PI3Kβ isoforms could be
developed on stratified cancer patient subpopulations based
on PIK3CA mutant or PTEN deficient genotypes, respectively.
INTRODUCTION
■
Phosphoinositide 3-kinase (PI3K) pathway activation plays a
major role in cancer as a result of abnormalities in major
components of this signaling cascade, including activating point
mutations and/or amplification of the PIK3CA gene as well as
loss of negative regulatory proteins such as Phosphatase and
TENsin homologue (PTEN).1 In addition, aberrant signaling
through this pathway has been shown to represent a key
mechanism of resistance to anticancer therapies including
targeted therapy. PI3K is considered as one of the most
promising target in this pathway for cancer treatment. Class I
PI3Ks are lipid kinases comprising four isoforms, PI3Kα,
PI3Kβ, PI3Kδ, and PI3Kγ, that generate phosphatidylinositol
3,4,5-triphosphate (PIP3), a second messenger involved in Akt
activation. Cellular PIP3 levels are tightly regulated by PTEN
lipid phosphatase, which converts PIP3 into phosphatidylino-
sitol 4,5-biphosphate (PIP2). The class I PI3Ks is furth,er
divided into class IA comprising PI3Kα, PI3Kβ and PI3Kδ,
which are activated by receptor tyrosine kinases or GPCR
(PI3Kβ) and subclass IB with a single member, PI3Kγ,
activated by GPCRs. Class IA proteins which consist of a
catalytic subunit (p110α, p110β, p110δ) and a regulatory
Received: February 23, 2012
Published: April 23, 2012
© 2012 American Chemical Society
4788
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
Figure 1. Structure of some PI3K inhibitors.
Furthermore, isoform-specific PI3K inhibitors may exhibit
better therapeutic windows compared to pan-PI3K inhibitors
and, depending on their pharmacological properties, could be
favorably combined with other targeted therapies or standard of
care agents.
and γ, respectively) in biochemical assays. Subsequent
medicinal chemistry efforts were devoted to the replacement
of the potentially labile anilide moiety to bioisosteric
benzimidazole or benzoxazole analogues. The unsubstituted
benzimidazole compound 615 (IC50 >10 μM, 158 nM, 2870
nM, and >10 μM on PI3Kα, β, δ, and γ, respectively) and
benzoxazole compound 715 (IC50 >10 μM, 82 nM, 779 nM,
and >10 μM on PI3Kα, β, δ, and γ, respectively) displayed a
remarkable level of PI3Kβ selectivity while maintaining high
ligand efficiency (LE) values16 of 0.40 and 0.42, respectively.
At the time we initiated this work, PI3Kβ specific inhibitors
derived from different chemotypes such as the quinolone TGX-
155,17 pyridopyrimidone TGX-221,18 and chromenone PIK-
10818 were reported in several publications. However, structural
information on their binding modes was limited and only an X-
ray sructure of the closely related pan-PI3K modulator LY-
294002 in complex with PI3Kγ was available.19 To determine
the binding mode of the pyrimidone benzimidazoles, we
attempted cocrystallization in the surrogate enzyme PI3Kγ, but
all efforts failed, probably due to their poor PI3Kγ binding
affinity (IC50 > 10 μM). To overcome this limitation, we
decided to embark on the thorough exploration of the
structure−activity relationship (SAR) of this series, introducing
additional substituents to improve biological activity and drug-
like properties.
To guide compound potency and selectivity improvement on
PI3Kβ, we generated a homology model of PI3Kβ based on
ligand-bound PI3Kγ X-ray structures available in-house and in
the public domain (PDB).19,20 Interestingly, the homology
model that best accommodated the compounds allowed the
identification of a region in the P-loop, made accessible by
movement of a conserved methionine (Met779), previously
described as the selectivity pocket (Met804 in PI3Kγ).20 In our
model, the docked compounds adopt a nonplanar geometry
where the morpholine makes a hydrogen bond with the hinge
backbone NH of Val854. The pyrimidone scaffold occupies the
central part of the ATP binding site from where the
benzimidazole side chain is projected into the selectivity pocket
Until very recently, relatively few isoform-selective PI3K
inhibitors, exhibiting appropriate drug-like properties, have
been described. CAL-101 was the first isoform-selective
compound entering clinical trials for the treatment of cancer.
On the basis of the restricted expression of PI3Kδ isoform in
hematopoietic cells, this PI3Kδ-selective compound is currently
being developed in hematologic cancer malignancies.7,8 phase 1
clinical trials with the PI3Kβ-selective inhibitors, AZD-6482
and TGX-155, were performed some years ago for thrombosis
but to the best of our knowledge these compounds were never
developed for cancer treatment. Very recently, PI3Kα-selective
inhibitors have entered clinical phase 1 for PIK3CA mutated
cancer patients.9,10 As part of our ongoing drug discovery and
development activities for targeted cancer therapies, we have
initiated the search for PI3K isoform-selective inhibitors. We
had previously reported the identification of PI3Kα/mTOR
and PI3Kα modulators that demonstrated some level of
antiproliferative selectivity in PIK3CA mutant versus PTEN-
deficient cancer cells.11 Similarly, we have also described
PI3Kβ-selective kinase inhibitors exhibiting superior cellular
activity in PTEN-deficient versus PIK3CA mutant tumor
cells.12
We report herein the discovery and optimization of a novel
series of PI3Kβ isoform-selective small molecular mass
inhibitors and their potential application as anticancer agents
in PTEN-deficient tumor models.
LEAD DISCOVERY AND OPTIMIZATION
■
Starting from a high-throughput screening campaign and
following early hit validation and chemical exploration, we
identified compound 513,14 (Figure 1), which presented some
interesting selectivity for PI3Kβ versus other PI3K isoforms
(IC50 = 2133 nM, 42 nM, 118 nM, and >10 μM on PI3Kα, β, δ,
4789
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
where it is stacking against the side chain of Trp787. Given
strong analogies with, on the one hand the morpholine
containing ligands, and on the other hand the “propeller
shaped”20 inhibitors such as PIK-39 (Figure 1), we confidently
adopted this binding mode hypothesis for analyzing SAR and
drive optimization of the pyrimidone series potency (see Figure
2).
pAkt (S473), the ratio of IC50 (T308)/(S473) varying from 0.3
to 2.5, which stands within assay variability.
Consequently, in the following Tables 2−5, we only report
pAkt (S473) inhibition in cells to document pharmacodynamic
modulation of the PI3K pathway.
The optimization of the benzimidazole (see Table 2) and
benzoxazole (see Table 3) series was performed in parallel to
exploit potential differences in the physicochemical properties
between these two structurally distinct chemotypes.
Analysis of the data presented in Table 2 showed that
introduction of a halogen (compounds 9, 13, and 14) brought
the level of inhibitory activity below 100 nM. Introduction of
the N-methyl benzimidazole compound 8 also led to a slight
improvement of activity, comparable to the combination of 5-
fluoro and N-methyl benzimidazole as in compound 11,
however, the 6-fluoro isomer 12 was less potent. Adding a
methyl group on the N−H of the pyrimidone scaffold as in
compound 10 led to a 15-fold loss in activity. This drastic drop
in potency suggested that polar groups interacting with the
residues Asp937 or Lys805 in PI3Kβ would be preferred at this
position. Contrary to the effect observed with halogens,
introduction of a methyl, an electron donating methoxy or an
electron withdrawing trifluoromethyl (compounds 15−17), did
not enhance activity. Substitution of the benzimidazole
nitrogen with groups larger than methyl such as ethyl,
cyclopropyl, methoxyethyl, cyclohexyl, phenyl, or benzyl
(compounds 18−23) led to some activity modulation albeit
with a loss of selectivity over PI3Kδ, especially for compounds
21−23, which were even more potent on PI3Kδ. Introduction
of the 4-phenyl (compound 24) led to a binding activity
increase of over 10-fold with PI3Kβ IC50 = 10nM, however,
with a similar selectivity over PI3Kδ. The N-methyl 4-phenyl
compound 25 and the N-methyl 7-phenyl compound 26 were
both equally potent, suggesting that there is room for the N-
methyl group to adopt both the inward and outward positions
in the selectivity pocket. The proposed binding mode was also
in agreement with the loss of activity observed with the 5-
phenyl and the 6-phenyl (compounds 27−28) because in these
positions the phenyl group would clash with residues in the P-
loop.
Figure 2. Proposed binding mode of pyrimidone benzimidazoles in
PI3Kβ showing favorable interactions with the hinge region (V854)
and the selectivity pocket (M779).
During this process, we examined the activity of these
compounds on the four PI3K isoforms in biochemical assays
and target modulation in tumor cells. To further confirm their
selectivity profile against other members of the lipid kinase
family, the new compounds were evaluated for mTOR
inhibition and shown to be all inactive (IC50s > 10 μM) in a
biochemical assay. Target modulation in tumor cells was
assessed by measuring the inhibition of Akt phosphorylation in
the PTEN-deficient PC3 prostate cancer cell line. We first
demonstrated for a set of molecules, including reference
compounds, that PI3Kβ-selective compounds inhibited Akt
phosphorylation at both T308 (phosphorylated by PDK1
kinase)21 and S473 (phosphorylated by mTORC2 kinase)21
residues with the same level of activity. As examplified in Table
1, reference compounds and pyrimidone benzimidazoles
exhibited roughly the same potency on pAkt (T308) and
Overall, we globally observed a relatively good relation
between the biochemical activity values and cellular potency for
the benzimidazole derivatives reported in Table 2. It is
interesting to point out that when plotting cellular potency
against biochemical activity on PI3Kβ, a trend for correlation
(r2 = 0.45, see Figure 3) was observed, suggesting an overall
modest cellular permeability. When removing the benzimida-
zole N−H compounds from this set, the correlation went up to
r2 = 0.67.
Table 1. Cellular Activity of Reference Compounds and
Pyrimidone Benzimidazoles on Akt Phosphorylation at
T308 and S473 Residues in PC3 Cell Line
In the benzoxazole series (see Table 3), a fluorine scan (29−
32) showed only a modest (2-fold) increase in potency for the
5-fluoro derivative compound 30.
cellular activity
Similar results were obtained in a methyl scan (compounds
33−37). In this case, a >10-fold drop in activity was observed
for the N-methyl pyrimidone derivative 33, which is consistent
with the data obtained with compound 10 in the benzimidazole
series. The synthetic intermediate 4-bromo 38 showed activity
comparable to the 4-fluoro 29 and the 4-methyl 34. The 4-
phenyl 39 showed a 4-fold increase in potency over the 7-
phenyl 40, and although there is no steric difference between
the two compounds, one possible explanation might lay in the
observation that Met779 is facing the azole part of the ligand
via its sulfur atom in a configuration stabilizing a N···S
biochemical
activity
a
a
a
PI3Kβ IC50 pAkt (T308) pAkt (S473)
IC50 ratio pAkt
(T308)/(S473)
compd
(nM)
IC50 (nM)
IC50 (nM)
TGX-
221
30
3
10
0.30
PIK-
108
28
3
3
1.00
8
99
30
188
100
76
41
2.47
2.44
9
a
IC50 values are reported as the mean from at least two independent
experiments; see Experimental Methods for assay details.
4790
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
Table 2. Biochemical and Cellular Activity of Pyrimidone Benzimidazoles (6, 8−26) and Reference Compounds (TGX-
221,TGX-155, and PIK-108)
a
a
a
a
a
PI3Kα IC50
PI3Kβ IC50
PI3Kδ IC50
PI3Kγ IC50
pAkt (S473) IC50
compd
R1
R2
R3
MW
(nM)
(nM)
(nM)
(nM)
(nM)
TGX-221
364
354
364
311
325
329
343
343
343
390
346
325
341
379
339
369
387
393
387
401
387
401
401
401
401
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
5460
30
32
851
166
658
2870
1395
219
5715
855
1060
442
247
2210
397
2850
225
36
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
8270
10
72
3
TGX-155
PIK-108
6
28
H
H
H
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
158
99
665
76
41
8
Me
H
9
5-F
30
10
H
5-F
739
82
11
Me
Me
H
5-F
16
289
85
12
6-F
290
91
13
5-Br
5-Cl
5-Me
5-OMe
5-CF3
H
14
H
5650
99
63
15
H
9880
450
193
386
72
140
475
961
26
16
H
>10000
>10000
1000
17
H
18
Et
19
cyclopropyl
−EtOMe
cyclohexyl
Ph
5-F
>10000
>10000
8400
29
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
9
20
H
205
90
270
21
109
142
8
21
H
22
H
6790
75
36
23
−CH2Ph
H
H
5725
111
10
84
105
65
24
4-Ph
4-Ph
7-Ph
5-Ph
6-Ph
5503
180
274
181
6748
3406
25
Me
314
57
26
26
Me
4575
60
17
27
Me
>10000
7798
>10000
3410
28
Me
a
IC50 values are reported as the mean from at least two independent experiments; see Experimental Methods for assay details.
nonbonded interaction between the C−S−C fragment and the
benzimidazole ring.22 Indeed, when searching the PDB
(Protein Data Bank) with the IsoStar knowledge base23
(version 2.1.2 tool package from CSD-Cambridge Structural
Database), 483 interactions between the methylthio moiety of a
methionine and an aromatic or sp2 N were found compared to
eight interactions with an aromatic oxygen (although this
search did not take into acount the methionine orientation).
Replacement of the 4-phenyl with the 4-thien-2-yl 41 or a 4-
pyridyl (42−44) did not lead to an improvement of affinity.
The best activity was obtained with the 4-phenyl-5-fluoro
derivative 45, which reached the single-digit nanomolar IC50
value in both biochemical and cellular assays, showing again the
beneficial effect of the 5-fluorine. When plotting the cellular
pAkt values against the biochemical activities on PI3Kβ, for the
benzoxazole series, we obtained a better correlation (r2 = 0.801,
see Figure 4) than the one observed for the benzimidazole
series, suggesting in particular a superior overall cell
penetration.
Potent benzimidazole and benzoxazole derivatives (PI3Kβ
IC50 < 100 nM and pAkt <500 nM) were further evaluated for
thermodynamic solubility at pH 7.424 and in vitro pharmaco-
kinetic properties including intestinal permeability, microsomal
stability, and CYP3A4 inhibition (see results in Table 4). All
selected compounds have molecular weight below 450 and
ClogP below 4.2, and none has any significant inhibitory
activity on recombinant human CYP3A4 (IC50 >40 μM except
for compound 41 = 19 μM)
Aqueous solubility on the other hand was surprisingly low,
considering the low molecular weight and low ClogP of these
compounds. We suspected strong crystal packing with
intermolecular hydrogen bonding and hydrophobic interac-
tions, especially for compounds 11, 13, 14, 24, 26, 30, 39, 40,
41, and 45 that have thermodynamic solubility less than 50 μM.
Melting points measured for some compounds were indeed
found to be high (>260 °C), which is in agreement with a
strong crystal packing. Permeability measurements in CaCo2
cell lines TC7 clone25 were found to be variable for
benzimidazoles and high for benzoxazoles, without any
correlation (r2= 0,146) with polar surface area (PSA). We
found that all N−H benzimidazoles (R2H) 9, 13, 14, and 24
had low permeability (<20 nm/s), and almost all N-substituted
benzimidazoles had better permeability except for 18 and 21,
The SAR generated for the benzimidazole and benzoxazole
series indicate that R1 substitution with methyl leads to
impairment of activity, while large groups such as phenyl at
positions 4 or 7 or halogens (fluorine) at position 5 provide
favorable interactions.
4791
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
Table 3. Biochemical and Cellular Activity of Pyrimidone Benzoxazoles (7, 29−45) Substitution
a
a
a
a
a
compd
R1
R2
MW
PI3Kα IC50 (nM)
PI3Kβ IC50 (nM)
PI3Kδ IC50 (nM)
PI3Kγ IC50 (nM)
pAkt IC50 (nM)
7
H
H
H
H
H
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
312
330
330
330
330
326
326
326
326
326
391
388
388
394
389
389
389
406
>10000
>10000
>10000
>10000
>10000
>10000
>10000
>10000
9768
82
567
46
779
8450
854
>10000
>10000
>10000
>10000
>10000
62
125
26
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
4-F
5-F
6-F
126
373
1333
461
960
526
425
585
19
9040
2555
49
7-F
88
H
4-Me
5-Me
6-Me
7-Me
4-Br
4-Ph
7-Ph
2755
3333
75
233
143
202
127
13
1137
2385
2271
373
>10000
>10000
>10000
>10000
>10000
5742
7150
>10000
7754
>10000
4357
90
1181
398
177
22
4-thiophen-2-yl
4-pyridin-2-yl
4-pyridin-3-yl
4-pyridin-4-yl
4-Ph; 5-F
46
3333
336
156
158
8
2170
335
>10000
>10000
>10000
>10000
113
44
>10000
>10000
>10000
995
47
151
6
a
IC50 values are reported as the mean from at least 2 independent experiment; see Experimental Methods for assay details.
Figure 3. PI3Kβ biochemistry pIC50 and pAkt cellular pIC50
correlation for the benzimidazoles 6−9 and 11−26.
Figure 4. PI3Kβ biochemistry pIC50 and pAkt cellular pIC50
correlation for the benzoxazoles 7−32 and 34−45.
R2Et and cyclohexyl, respectively. To elucidate the low
permeability of the benzimidazoles, we assessed transport in
both apical to basolateral (A−B) and basolateral to apical (B−
A) directions across the cell monolayer, enabling the efflux ratio
determination which was found to be high (>10). Benzox-
azoles, on the other hand, displayed lower efflux ratio (<5).
Overall, we could observe that low permeability was linked to
strong efflux ratio, most likely due to P-gp interaction.
Metabolic lability measurements were found to be high
(>40%) for all compounds except 8, 11, and 45 in human
liver microsomes and also for half of the compounds (18, 19,
21, 22, 24−26, 40, and 41) in mouse liver microsomes. As
illustrated in the heat map (Figure 5), compound 8 was the
only molecule with a well balanced in vitro pharmacological
profile and was thus selected for further structural and
biological characterization.
The protein kinase selectivity for this compound was
assessed with a Caliper-based assay26 on a 192-kinases panel.
Remarkably, the compound was found highly selective with no
activity up to 10 μM on all kinases tested, including mTOR
(data not shown).
X-Ray Structure of Compound 8 in PI3Kδ. At the time
of this work, PI3Kγ was the only PI3K isoform for which X-ray
structures of protein−ligand complexes could be readily
obtained. However, despite numerous attempts, no ligand−
protein complex structures were successfully obtained with the
pyrimidone benzimidazole series in PI3Kγ, probably due to its
4792
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
Table 4. In Vitro and Calculated Properties of Active Compounds 7−9, 11, 13, 14, 18, 19, 21, 24−26, 30, 39−41, and 45
a
a
b
c
d
PI3Kβ IC50
pAkt (S473) IC50
solubility pH
7.4 μM
CaCo2 Papp
(nm/s)
MLM
HLM
rhCYP3A4 IC50
compd MW
(nM)
(nM)
ClogP PSA
% lability
% lability
μM
7
312
325
329
343
390
346
339
369
393
387
387
401
401
330
388
388
432
406
82
76
30
82
91
99
72
29
90
75
10
57
60
46
19
90
46
8
62
130
41
150
85
63
26
9
1.8
2
80
72
83
72
83
83
72
72
72
72
83
72
72
80
80
80
108
80
160
65
76
3
199
30
2
7
13
0
41
27
69
39
75
78
60
99
97
98
78
41
57
47
40
57
60
>40
>40
>40
>40
>40
>40
>40
>40
>40
>40
>40
>40
>40
>40
8
9
1.7
2
11
13
14
18
19
21
22
24
25
26
30
39
40
41
45
31
0
20
27
35
48
99
96
100
51
40
55
23
26
61
79
17
2.4
1.9
2.5
2.4
4.1
3.9
3.2
3.8
3.8
1.9
3.6
3.6
3.4
4.1
3
3
0
221
5
22
15
44
1
142
8
397
3
65
26
17
26
13
177
22
6
114
39
160
310
323
255
371
2
3
3
3
>40
19
3
15
18
>40
a
b
IC50 values are reported as the mean from at least 2 independent experiments, see Experimental Methods for assay details. Mouse liver microsome.
c
d
Human liver microsome. Recombinant human CYP3A4.
Figure 6. Binding of compound 8 to the ATP binding site of p110δ
(see Table 8 for X-ray data collection and refinement).
Figure 5. Heat map of Table 4 properties with selected thresholds
(with the color coding: green, pass; red, fail).
for LY294002, and several other previously reported20,25−27
morpholine containing compounds binding to PI3Kγ or PI3Kδ.
In the P-loop, Met752 changed conformation relative to its
position in the PI3Kδ apo-structure, leaving space for the
methyl-benzimidazole moiety to bind in the specificity pocket.
The (R2) methyl group points toward residues at the entrance
of the ATP binding site (Asp832, Thr833, and Asn836),
however, leaving some space for larger substituents in PI3Kδ.
Interestingly, superposition between the X-ray structure of the
PI3Kδ-specific compound PIK-39 and the X-ray structure of
compound 8 corroborated the observed SAR for the
pyrimidone benzimidazole series. As exemplified by com-
pounds 21, 22, and 23, there is a marked increase in PI3Kδ
potency over the other three isoforms when adding large
substituents, comparable to the 2-methoxy phenyl in PIK-39, to
the R2 position. The N−H of the amide in the pyrimidone
limited affinity for this isoform. Recently, the X-ray structure of
PI3Kδ was described in its apo-form as well as in complex with
ligands displaying diverse selectivity profiles, including
compounds with PI3Kβ inhibition.27 To compensate the lack
of a PI3Kβ X-ray structure, and encouraged by measurable
potency of our compounds on the PI3Kδ isoform, we set about
to solve the structure of PI3Kδ in complex with the pyrimidone
series. The structure of compound 8 (IC50 of 1395 nM on
PI3Kδ) was solved in PI3Kδ (Figure 6), revealing a binding
mode in agreement with the binding mode hypothesis in PI3Kβ
used for pyrimidone benzimidazole optimization. The morpho-
line moiety bound the hinge region via the main-chain nitrogen
of Val828 in a very similar manner to what had been described
4793
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
scaffold is pointing toward a partly solvent exposed region lined
with Asp911 and Lys779, further supporting the hypothesis that
small polar groups would be preferred over a methyl in this
position. The original goal of this crystallography effort was to
understand the structural basis for PI3K isoform specificity and
to use the information to design more selective PI3Kβ
inhibitors. However, from the structure of compound 8
bound to PI3Kδ, one can observe that the large majority of
residues in close contact with the inhibitor were conserved
within the four isoforms. No specific interaction could explain
on the one hand the pronounced affinity of the pyrimidone
series for PI3Kβ and PI3Kδ isoforms compared to PI3Kα or
PI3Kγ and on the other hand the observed PI3Kβ specificity for
compounds of these series. It has previously been suggested20,28
that differences in protein plasticity, particularly in the P-loop,
would bring PI3Kδ specificity for compounds occupying the
specificity pocket. On the basis of the structural results for a
PI3Kβ specific compound reported here, supported by the
close sequence homology between PI3Kβ and PI3Kδ relative to
PI3Kγ or PI3Kα,29 we propose to extend this hypothesis by
suggesting that not only PI3Kδ but also PI3Kβ can readily
adopt a conformation in which the specificity pocket is
accessible for productive ligand binding. A conformation that
would be energetically disfavored in the PI3Kα and PI3Kγ
isoforms. To determine the veracity of this hypothesis and to
clarify the origins of PI3Kβ versus PI3Kδ specificity, further
structural as well as thermodynamic studies of the class I PI3K
isoforms and especially the p110β, for which the first structure
was recently solved,30 would be of great value.
Cellular Selectivity. Activity of compound 8 on the four
PI3K isoforms was further evaluated in cellular settings. To this
end, the following cell-based assays, measuring Akt phosphor-
ylation inhibition, were used to assess each isoform, PIK3CA-
mutated H460 lung cancer cells for PI3Kα, PTEN-deficient
PC3 prostate cancer cells or MEF-3T3-myr p110β mouse
fibroblasts overexpressing activated p110β for PI3Kβ, MEF-
3T3-myr p110δ mouse fibroblasts overexpressing p110δ for
PI3Kδ and Raw264.7 mouse macrophages (after stimulation of
Akt phosphorylayion by C5a) for PI3Kγ (Table 5).
Under these experimental conditions, compound 8 was
shown to be at least 100-fold more potent on PI3Kβ than on
PI3Kα and PI3Kγ isoforms and above 10-fold more potent over
PI3Kδ. In comparison, the reference compound TGX-221
exhibited roughly the same level of PI3Kβ selectivity versus
other isoforms. Compound 8 was shown to exhibit selectivity in
cellular settings in agreement with what was observed in
biochemical assays.
In Vivo Studies. Compound 8 was further evaluated in a
pharmacokinetic (PK) study in female CB17/lCR-Prkdc severe
combined immunodeficiency (SCID)/Crl mice showing an oral
bioavailability of 26% (Table 6B). Pharmacokinetic parameters
from the intravenous (iv) route (Table 6A) indicated a
moderate clearance (3.1 L/h/kg) relative to hepatic blood flow
(5.2 L/h/kg), a moderate Vdss (1.8 L/kg), and a long terminal
elimination half-life (T1/2 = 4.4 h).
Figure 7 illustrates that compound 8 has comparable
exposure when administered at 6 mg/kg iv in solution and at
10 mg/kg po in suspension and similar terminal halflife.
Given the clear potential of compound 8 to selectively inhibit
PI3Kβ in PTEN-deficient tumor cells and its favorable
pharmacokinetic properties, its in vivo activity was evaluated
in a mouse pharmacodynamic (PD) experiment. Compound 8
was administered orally at 300 mg/kg in SCID mice bearing
human PTEN-deficient PC3 prostate carcinoma tumors, and a
kinetic PK/PD test was performed harvesting tumor and
plasma samples at 0.5, 1, 3, 6, and 24 h post dose. Target
modulation in tumor tissue was assessed by measuring Akt
phosphorylation on T308 residue. Figure 8A illustrates pAkt
(T308) (bar graph, left axis) and the observed plasma
concentrations of compound 8 (blue curve, right axis) at
each time point. Tumor concentrations were also measured in
this study and were found to be 1−3-fold lower than plasma
concentrations (Table 7).
This experiment revealed a sustained and potent target
inhibition induced by treatment with compound 8, with more
than 50% pAkt (T308) inhibition for at least 6 h (Figure 8a).
This effect did correlate directly with compound exposure. A
curve of plasma or tumor concentration versus percent
inhibition of Akt phosphorylation is shown in Figure 8B and
revealed a direct PK-PD response model. The in vivo effective
concentration giving 50% PD modulation (EC50) calculated
from different experiments was estimated to be around 2 μM
(1.6 μM for plasma concentrations and 2.1 μM for tumor
concentrations).
A kinetic PK/PD experiment was also performed after
repeated administration of compound 8 for four days with two
different dosing schedules, qd or bid. This experiment
demonstrated the maintenance of pAkt (T308) inhibition
over 24 h with the bid dosing at 300 mg/kg. Plasma
concentrations at day 1 and day 4 were comparable at the
different time points tested, indicating that there was neither
compound accumulation due to inhibition of its own
metabolism nor clearance related to autoinduction over four
days (data not shown). This experiment supported the bid
scheduling for the maintenance of continuous inhibition of Akt
phosphorylation.
Table 5. Biochemical and Cellular PI3Kβ-Selectivity vs
Other Isoforms: (A) Compound 8, (B) TGX-221
(A)
b
compd 8 IC50 (nM)
PI3Kα
PI3Kβ
PI3Kδ
PI3Kγ
biochemistry
cellular
>10000
99 (LE = 0.40)
1395
>10000
>10000
a
a
a
a
>10000
76
a
2655
To determine the functional consequence of PI3Kβ
inhibition, we then investigated the antitumor activity of
compound 8 in human PTEN-deficient PC3 prostate
carcinoma tumor model subcutaneously xenografted in SCID
mice. In this study, tumors were allowed to reach at least 100
mm3 before compound or vehicle (control curve) admin-
istration, and tumor volume was measured regularly over the
treatment period. Administered orally at 300 mg/kg bid for 9
days in PC3 tumor-bearing mice, compound 8 reduced tumor
growth with a ΔT/ΔC = 45% on day 32 at the end of the
treatments (p = 0.0214) (Figure 9).
86
(B)
b
TGX-221 IC50 (nM)
PI3Kα
PI3Kβ
30
PI3Kδ
851
PI3Kγ
biochemistry
cellular
>10000
>10000
a
a
a
a
>1000
10
a
252
>1000
11
a
pAkt-S473 was measured by MSD technology in H460, PC3, MEF-
3T3-Myr p110β, MEF-3T3-Myr p110δ, and Raw264.7 cells,
b
respectively. IC50 values are reported as the mean from at least two
independent experiments, see Experimental Methods for assay details.
4794
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
Table 6. Pharmacokinetic Parameters of Compound 8 in Female SCID Mice: (A) After a Single Intravenous Administration of 6
mg/kg in Solution in 10% Glycofurol, 5% PS80 in Glucose 5%, (B) After a Single Oral Administration of 10 mg/kg in
Suspension in 0.6% Methylcellulose, 0.5% Tween 80
(A)
compd 8 iv (6 mg/kg)
C0 (ng/mL)
AUC(0−Tlast) (h·ng/mL)
Tlast (h)
AUC(0−24 h) (h·ng/mL)
AUC(0−inf) (h·ng/mL)
T1/2 (h)
Cl (L/h/kg)
3.1
Vdss (L/h/kg)
3.1
4870
1950
24
1950
(B)
1950
4.4
compd 8 po (10 mg/kg in suspension)
Cmax (ng/mL)
Tmax (h)
AUC(0−Tlast) (h·ng/mL)
Tlast (h)
AUC(0−24 h) (h·ng/mL)
AUC(0−inf) (h·ng/mL)
T1/2 (h)
3.9
F (%)
641
0.5
846
24
846
850
26
Table 7. PK Exposure from Compound 8 PK/PD Study in
PC3 Mice Xenograft Model
dose 300 mg/kg po
time (h)
0.5
28
13
1
3
8
6
6
24
1.7
0.7
plasma (μM)
tumor (μM)
22
16
11
3.6
shRNA against PIK3CB led to a comparable tumor growth
delay in PC3 model.
CHEMISTRY
■
As shown in Scheme 1, the 2-[4-(morpholin-4-yl)-6-oxo-1,6-
dihydropyrimidin-2-yl]-N-phenylacetamide 513 can be obtained
from the carboxylate 2 by condensation of aniline in the
presence of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
(EDCI) and pyridine in dimethylformamide (DMF). The
novel intermediate 113 was efficiently obtained by a one-pot
reaction between morpholine and an excess (superior to 2
equiv) of commercially available ethyl 3-amino-3-ethoxyacrylate
hydrochloride in the presence of N,N-diisopropylethylamine
(DIPEA) in ethanol under reflux. Reactions reported in the
literature use stoichiometric amount of morpholine and ethyl 3-
amino-3-ethoxyacrylate hydrochloride and were intended to
Figure 7. Time course exposure (logarithmic scale) of compound 8
after a single intravenous and oral administration to female SCID mice
(arithmetic means for plasma, n = 3).
In this study, treatment with compound 8 induced tumor
growth delay at well tolerated doses, with no sign of toxicity
and no body weight loss. These results are in agreement with
the published data in which downregulation of PIK3CB using a
shRNA leads to PI3K pathway modulation and tumor growth
inhibition selectively in PTEN-deficient genetic context
(reported T/C = 49%).5 Remarkably, compound 8 and
Figure 8. Effect of compound 8 on pAkt T308 inhibition. (A) In SCID mice PC3 xenograft model at 300 mg/kg po in suspension (percentage of
pAkt-T308 inhibition on the left Y axis (bars reported represent the mean percentage and standard error of the mean of 3 mice per group) and
plasma concentration [C] in μM log scale on the right Y axis). (B) EC50 measured by plotting concentration (plasma and tumor) vs inhibitory effect
on pAkt-T308. E = EmaxC/EC50 + C, where E denotes compound 8 effect on biomarker; Emax, the maximal effect of the compound on biomarker; C,
compound 8 concentration in μM; EC50, the plasma or tumor concentration corresponding to the half-maximal response of the biomarker pAKT-
T308.
4795
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
acid at reflux led to the expected benzimidazoles in variable
yields (see Experimental Methods for details).
Alternatively, the benzimidazoles 11, 12, 18, 25, and 26 were
isolated after alkylation of the benzimidazole moiety with alkyl
halides in the presence of sodium hydroxide in acetone
(Scheme 3).
As shown in Scheme 4, the benzoxazoles 7, 29−38, 40, 42,
and 45 were synthesized in two steps. The condensation of the
carboxylates 2 and 4 with the corresponding substituted
aminophenols followed by an intramolecular cyclization in o-
xylene in the presence of p-toluenesulfonic acid (PTSA) at
reflux led to the expected benzoxazole compounds in variable
yields (see Experimental Methods for details).
Scheme 5 shows how the substituted benzoxazoles 39, 41,
43, and 44 were synthesized from 38 by a Suzuki cross-
coupling reaction with the corresponding boronic acids (or
esters). The reactions were performed in 1,2-dimethoxyethane
(DME) and water in the presence of sodium carbonate and
tetrakis (triphenylphosphine) palladium(0) (Pd(PPh3)4) cata-
lyst under argon atmosphere. The corresponding substituted
benzoxazoles were obtained in moderate yield after column
chromatography.
Figure 9. Compound 8 antitumor efficacy in mice PC3 xenograft
model.
stop at the amidine intermediate (between brackets in Scheme
1) which can be isolated as the hydrochloride.31−35 In our case,
the use of 3 equiv of imidate combined with the introduction of
DIPEA led to the isolation of the novel ester 1, formed most
likely through condensation of the intermediate amidine with a
second equivalent of imidate. Compound 1 was converted into
3 via methylation with methyl iodide in dioxane in the presence
of cesium carbonate at 40 °C under argon. The two sodium
salts 2 and 4 were obtained at room temperature by hydrolysis
of the esters 1 and 3, respectively, in tetrahydrofuran with 1
equiv of sodium hydroxide. The acids were isolated and
conserved as the sodium salt because it was noticed that the
acid has the tendency to decarboxylate even a room
temperature.
CONCLUSION
■
We have discovered a new series of high LE benzimidazole and
benzoxazole pyrimidones as potent and selective PI3Kβ
inhibitors which have been prepared by a new short and
efficient synthesis, utilizing the novel intermediate compound 1.
Optimization of the series, through SAR analysis, physicochem-
ical and in vitro pharmacokinetic optimization has led to the
identification of compound 8. Crystallized in PI3Kδ, this
molecule takes a conformation likely equivalent to the one it
could adopt in PI3Kβ and that sheds light on important
features in pyrimidone series SAR and on PI3Kβ isoform
specificity. Compound 8 was found to be a potent and selective
PI3Kβ inhibitor in biochemical and cellular settings. It
demonstrated good selectivity versus other class I PI3K
isoforms but also versus a large panel of protein kinases. This
molecule was shown to potently inhibit Akt phosphorylation in
The benzimidazoles 6, 8−10, 13−17, 19−24, and 27−28
were synthesized in two steps (Scheme 2). The condensation of
the carboxylates 2 and 4 with the corresponding substituted
diamines followed by an intramolecular cyclization in acetic
a
Scheme 1. Synthesis of Compound 5 and Intermediates 1−4
a
Reagents and conditions: (a) DIPEA, EtOH, reflux, 24 h, 46%; (b) MeI, Cs2CO3, dioxane, 40 °C, 16 h, 75%; (c) NaOH, THF, rt, 48 h, 100%; (d)
EDCI, pyridine, DMF, rt, 16 h, 70%.
4796
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
a
Scheme 2. Synthesis of Benzimidazoles 6, 8−10, 13−17, 19−24, 27, and 28 from Table 1
a
Reagents and conditions: (a) EDCI, pyridine, DMF, rt, 16 h; (b) AcOH, reflux, 4 h, overall yield 4−66%
a
Scheme 3. Preparation of Benzimidazoles 11, 12, 18, 25, and 26
a
Reagents and conditions: (a) NaOH, Acetone, rt, 16 h.
a
Scheme 4. Synthesis of Benzoxazoles 7, 29−38, 40, 42, and 45 from Table 2
a
Reagents and conditions: (a) EDCI, pyridine, DMF, rt, 16 h; (b) PTSA, o-xylene, reflux, 5 h, overall yield 2  67%.
a
Scheme 5. Preparation of Benzoxazoles 39, 41, 43, and 44
a
Reagents and conditions: (a) ArB(OH)2, Pd(PPh3)4, Na2CO3, DME, H2O, 100 °C, 5 h, yield 15−53%
4797
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
a
Scheme 6. Preparation of 2-Amino-6-fluorobiphenyl-3-ol
a
Reagents and conditions: (a) (NH4)2Ni(SO4)2·6H2O, HNO3, rt, 20 min. 31%; (b) Pd(dppf)Cl2, Cs2CO3 dioxane/water, 95 °C, 5 h, 88%; (c) Fe,
AcOH, 70 °C, 30 min. 78%.
PTEN-deficient PC3 prostate carcinoma cell line. In addition, it
proved to exhibit good PK properties, allowing sustained PI3K
pathway modulation and tumor growth delay in PTEN-
deficient tumor xenografts. The biological profile of this
molecule validated the benzimidazole and benzoxazole
pyrimidone series as small molecules able to slow down the
growth of PTEN-deficient tumors in vivo by targeting PI3Kβ
kinase activity selectively. These results have a strong impact for
the ongoing efforts to discover drugs targeting the PI3K
pathway with new profiles. Isoform-selective compounds
targeting only the relevant PI3K isoform for a given genetic
context should avoid toxicities linked to unnecessary target
inhibition and should provide a larger therapeutic index. It is
certainly conceivable that the dependence of PTEN-deficient
tumors on PI3Kβ is highly dependent on the presence of other
genetic abnormalities and that this approach will have to be
combined with other targeted therapies to reach tumor
regressions in a more complex genetic context encountered
in the majority of human cancer diseases.
Table 8. X-Ray Data Collection and Refinement Statistics for
Compound 8
a
Data Collection
space group
unit cell dimensions
a, b, c (Å)
P21
63.4, 220.5, 78.3
90, 113.9, 90
71.57−2.80 (2.87−2.80)
157600
α, β, γ (deg)
resolution (Å)
total reflections
unique reflections
completeness (%)
Rsym
47547
98.4 (99.5)
7.3 (47.1)
I/σ(I)
10.1 (2.4)
Refinement
reflections used
47377
25.2/27.7
74.8
R/Rfree
average B-values (Å2)
no. of atoms
protein
12514
82
ligand
EXPERIMENTAL METHODS
Chemistry. The nomenclature of the compounds was performed
using ACDLABS software, Version 11.01.
■
solvent
34
rms deviations
bond lengths (Å)
bond angles (deg)
0.007
0.87
All solvents and reagents obtained from commercial sources were
used without further purification.
Thin layer chromatography was carried out on Merck silica gel 60
F254 glass plates. Flash chromatography was performed using
prepacked Merck silica gel cartridges (15−40 μm).
The microwave apparatus used is a Biotage, Initiator 2.0, 400 W
max, 2450 MHz instrument.
a
Data in parentheses correspond to the highest resolution shells. To
calculate Rfree, 5% of the reflections were excluded from the
refinement. Rsym was defined as Rsym = ∑hkl ∑i|Ii(hkl) − ⟨I(hkl)⟩|/
∑
∑i Ii(hkl). rms, root mean square deviation.
hkl
The 1H NMR spectra at 400 MHz and the 1H NMR spectra at 500
MHz were performed on a Bruker Avance DRX-400 or Bruker Avance
DPX-500 spectrometer with the chemical shifts (δ in ppm) in the
solvent dimethyl sulfoxide-d6 (DMSO-d6) referenced at 2.5 ppm at a
temperature of 303 K and coupling constants (J) were given in hertz.
The mass spectra (MS) were obtained by methods A−C:
Method A. Waters UPLC-SQD instrument; ionization, positive
and/or negative mode electrospray (ES ). Chromatographic con-
ditions: column, Acquity BEH C18 1.7 μm, 2.1 mm × 50 mm.
Solvents: A, H2O (0.1% formic acid); B, CH3CN (0.1% formic acid);
column temperature, 50 °C; flow rate, 1 mL/min; gradient (2 min),
from 5 to 50% of B in 0.8 min; 1.2 min, 100% of B; 1.85 min, 100% of
B; 1.95, 5% of B; retention time = Tr (min).
1
Purities for final compounds were measured using H NMR or LC
(UV detection at 220 nm) and are >95% unless otherwise noted.
Ethyl [4-(Morpholin-4-yl)-6-oxo-1,6-dihydropyrimidin-2-yl]-
acetate (1). Ethyl 3-ethoxy-3-iminopropanoate hydrochloride (168.5
g (864 mmol)) and 155 mL (891 mmol) of N,N-diisopropylethyl-
amine in 200 mL of ethanol were added to a solution of 25 g (287
mmol) of morpholine in 400 mL of ethanol heated to 95 °C. The
reaction mixture was heated at 95 °C for 30 h and then allowed to
return to ambient temperature. The precipitate formed was filtered off
through sintered glass and then washed with 100 mL of ethanol, twice
500 mL of water, and finally, 500 mL of diethylether. The solid was
1
dried under vacuum to give 1 (35 g, 46%) as a white solid. H NMR
(400 MHz): 1,19 (t, J = 7.1 Hz, 3 H), 3.38−3.44 (m, 4 H), 3.56 (s,
2H), 3.61 (dd, J = 4.0 and 5.7 Hz, 4 H), 4.12 (q, J = 7.1 Hz, 2 H), 5.20
(s, 1 H), 11.69 (br s, 1 H). Mass spectrometry: method A, Tr (min) =
0.48; [M + H]+, m/z 268; [M − H]−, m/z 266.
Sodium [4-(Morpholin-4-yl)-6-oxo-1,6-dihydropyrimidin-2-
yl]acetate (2). Sodium hydroxide (18.7 mL (1 equiv), 2 M) were
added to a solution of 10 g (37 mmol) of 1 in 300 mL of
tetrahydrofuran. The reaction mixture was stirred for 48 h at ambient
temperature. The precipitate formed was filtered off through sintered
glass, washed with ethyl acetate, and rinsed several times with diethyl
ether. The solid obtained was then dried under vacuum to give 2 (8.7
g, 89%) as a white solid. 1H NMR (400 MHz): 3.08 (s, 2 H), 3.38 (t, J
= 4.6 Hz, 4 H), 3.61 (t, J = 4.6 Hz, 4 H), 5.08 (s, 1 H), 13.16 (br s, 1
Method B. Waters ZQ instrument; ionization, positive and/or
negative mode electrospray (ES ). Chromatographic conditions:
column, XBridge C18 2.5 μm, 3 mm × 50 mm. Solvents: A, H2O (0.1%
formic acid); B, CH3CN (0.1% formic acid); column temperature, 70
°C; flow rate, 0.9 mL/min; gradient (7 min), from 5% to 100% of B in
5.3 min; 5.5 min, 100% of B; 6.3 min, 5% of B; retention time = Tr
(min).
Method C. Waters UPLC instrument, BEH C18 column, 2.1 mm ×
50 mm; 1.7 μ, H2O + 0.1%TFA:AcN + 0.08% TFA 95:5 (0 min) to
5:95 (1.1 min) to 5:95 (1.7 min) to 95:5 (1.8 min) to 95:5 (2 min),
0.9 mL/min 55 °C; Waters SQD single quadrupol, 0.5 s scan time for
mass 120−1200; retention time = Tr (min).
4798
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
H). MS: method A, Tr (min) = 0.29; [M + H]+, m/z 240; [M − H]−,
m/z 238.
mixture was brought to reflux overnight. After cooling to room
temperature, the insoluble material formed was filtered off. The filtrate
was concentrated to dryness under reduced pressure, and the residue
was then purified by flash chromatography, eluent dichloromethane/
methanol, 98/02 then 90/10 to give 7 (30 mg, 8%) as a white solid;
mp 224 °C. 1H NMR (400 MHz): 3.35 (t, J = 5.0 Hz, 4 H), 3.56 (t, J
= 4.9 Hz, 4 H), 4.25 (s, 2 H), 5.25 (br s, 1 H), 7.31−7.40 (m, 2 H),
7.62−7.74 (m, 2 H), 11.92 (br s, 1 H). MS: method A, Tr (min) =
0.60; [M + H]+, m/z 313; [M − H]−, m/z 311.
Ethyl [1-Methyl-4-(morpholin-4-yl)-6-oxo-1,6-dihydropyri-
midin-2-yl]acetate (3). To a suspension of 500 mg (1.9 mmol) of
1 in 15 mL of dioxane were added 330 mg (2.4 mmol) of potassium
carbonate and 0.15 mL (2.3 mmol) of methyl iodide. The reaction
mixture was heated at 40 °C for 16 h and then cooled to ambient
temperature. The suspension was filtered through sintered glass and
then rinsed with dioxane, and the filtrate was concentrated under
reduced pressure. The residue was purified by silica column
chromatography, elution being carried out with a mixture of
dichloromethane, acetonitrile, and methanol (98/01/01, 96/02/02,
2-[(1-Methyl-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (8). By a similar procedure to that described
for the synthesis of 6 using 3 g (11.5 mmol) of 2 and 2.81 g (23
mmol) of N-methyl-1,2-phenylenediamine, 8 (2.05 g, 55%) was
1
then 90/05/05 V/V/V) to give 3 (200 mg, 38%) as a white solid. H
1
NMR (400 MHz): 1.21 (t, J = 7.2 Hz, 3 H), 3.28 (s, 3H), 3.40 (m, 4
H), 3.61 (m, 4 H), 3.92 (s, 2H), 3.61 (q, J = 7.2 Hz, 2H), 5.35 (s, 1
H). MS: method A, Tr (min) = 0.53; [M + H]+, m/z 282; [M − H]−,
m/z 280.
obtained a white solid; mp 257 °C. H NMR (400 MHz): 3.34 (m, 4
H), 3.56 (m, 4 H), 3.80 (s, 3 H), 4.20 (s, 2 H), 5.21 (s, 1 H), 7.17 (t, J
= 7.8 Hz, 1 H), 7.23 (t, J = 7.8 Hz, 1 H), 7.52 (d, J = 7.8 Hz, 1 H),
7.56 (d, J = 7.8 Hz, 1 H), 11.85 (br s, 1 H). MS: method A, Tr (min) =
0.36; [M + H]+, m/z 326; [M − H]−, m/z 324.
2-[(5-Fluoro-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (9). By a similar procedure to that described
for the synthesis of 6 using 7.00 g (26.8 mmol) of 2 and 6.76 g (53.6
mmol) of 1,2-diamino-4-fluorobenzene, 9 (3.00 g, 34%) was obtained
as an off-white solid. 1H NMR (400 MHz): 3.37 (t, J = 4.9 Hz, 4 H),
3.58 (t, J = 4.9 Hz, 4 H), 4.08(s, 2 H), 5.22 (s, 1 H), 7.00 (t, J = 8.2
Hz, 1 H), 7.31 (dd, J = 0.7 and 9.8 Hz, 1 H), 7.49 (br s, 1 H), 12.15
(br s, 2 H). MS: method A Tr (min) = 0.41; [M + H]+, m/z 330; [M
− H]−, m/z 328.
Sodium [1-Methyl-4-(morpholin-4-yl)-6-oxo-1,6-dihydropyr-
imidin-2-yl]acetate (4). Sodium hydroxide (13.74 mL (1 equiv), 2
M) were added to a solution of 7.73 g (27.5 mmol) of 2 in 200 mL of
tetrahydrofuran. The reaction mixture was stirred for 48 h at ambient
temperature. The precipitate formed was filtered through sintered
glass, washed with ethyl acetate, and rinsed several times with diethyl
ether. The solid obtained was then dried under vacuum to give 4 (7.56
g; 100%) as a white solid. 1H NMR (400 MHz): 3.29−3.33 (m,
partially masked, 5 H), 3.38 (t, J = 4.7 Hz, 4 H), 3.61 (t, J = 4.7 Hz, 4
H), 5.24 (s, 1 H). MS: method B, Tr (min) = 1.67; [M + H]+, m/z
254; [M − H]−, m/z 252.
2-[4-(Morpholin-4-yl)-6-oxo-1,6-dihydropyrimidin-2-yl]-N-
phenylacetamide (5). 2 (1.04 g (4 mmol)) were introduced into 16
mL of dimethylformamide, and then 0.65 mL (8 mmol) of pyridine,
1.0 g (5.2 mmol) of N-[3-(dimethylamino)propyl]-N′-ethylcarbodii-
mide hydrochloride (EDCI), and 0.75 g (8.1 mmol) of aniline were
added. The reaction mixture was stirred at ambient temperature
overnight and then evaporated to dryness under reduced pressure.
Water and ethyl acetate were added, and the resulting mixture was
thus stirred for 30 min. The precipitate formed was filtered off, washed
with diethyl ether, and dried to give 5 (879 mg, 70%) as a beige solid.
1H NMR (400 MHz): 3.38−3.44 (m, 4 H), 3.56−3.63 (m, 6 H), 5.20
2-[(5-Fluoro-1H-benzimidazol-2-yl)methyl]-3-methyl-6-
(morpholin-4-yl)pyrimidin-4(3H)-one (10). By a similar procedure
to that described for the synthesis of 6 using 400 mg (1.5 mmol) of 4
and 367 mg (2.9 mmol) of 1,2-diamino-4-fluorobenzene, 10 (75 mg,
1
15%) was obtained as an off-white solid, mp 265 °C. H NMR (400
MHz): 3.36 (m, 4 H), 3.37 (s, 3 H), 3.57 (m, 4 H), 4.37 (s, 2 H), 5.37
(s, 1 H), 7.00 (dt, J = 2.0 and 9.0 Hz, 1 H), 7.31 (dd, J = 2.0 and 10.5
Hz, 1 H), 7.50 (dd, J = 4.4 and 9.0 Hz, 1 H), 12.43 (br m, 1 H).
2-[(5-Fluoro-1-methyl-1H-benzimidazol-2-yl)methyl]-6-
(morpholin-4-yl)pyrimidin-4(3H)-one (11) and 2-[(6-Fluoro-1-
methyl-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-yl)-
pyrimidin-4(3H)-one (12). Sodium hydroxide (2N) (0.75 mL (1.5
mmol)) was added to a solution of 329 mg (1 mmol) of 9 in 15 mL of
acetone. After stirring at room temperature for 10 min, 93 μL (1.5
mmol) of iodomethane were added and the reaction mixture was
stirred overnight. The solvent was evaporated to dryness under
reduced pressure, the residue was diluted in 20 mL of water, and the
medium was brought to pH 6 using hydrochloric acid (2N). After the
insoluble material has been filtered off, the filtrate was evaporated to
dryness under reduced pressure and purified by flash chromatography;
eluent, dichloromethane/methanol 95/05 to give a 50/50 mixture of
two isomers (145 mg, 42%). The two isomers were separated by chiral
chromatography on Chiralpak AD 20 μm (1000 g, diameter 80 mm;
mobile phase, ethanol/heptane/triethylamine 60/40/0.1; mobile
phase flow rate, 150 mL/min; UV detection, 240 nm) to give 11
(s, 1H), 7.06 (t, J = 7.8 Hz, 1 H), 7.31 (t, J = 8.6 Hz, 2 H), 7.56 (d, J =
8.6 Hz, 2 H), 10.14 (s, 1 H), 11.64 (br s, 1H). MS: method A, Tr
(min) = 0.55; [M + H]+, m/z 315; [M − H]−, m/z 313.
2-(1H-Benzimidazol-2-ylmethyl)-6-(morpholin-4-yl)-
pyrimidin-4(3H)-one (6). To a suspension of 300 mg (1.1 mmol) of
2 in a mixture of 2.5 mL of dimethylformamide and 2.5 mL of
pyridine, 330 mg (1.7 mmol) of N-[3-(dimethylamino)propyl]-N′-
ethylcarbodiimide hydrochloride and 248 mg (2.3 mmol) of benzene-
1,2-diamine were added. The reaction mixture was stirred at ambient
temperature overnight and then evaporated to dryness under reduced
pressure. The residue was taken up in 5 mL of acetic acid and brought
to reflux for 1 h and then concentrated under reduced pressure. Then
100 mL of water were added and a saturated aqueous solution of
sodium bicarbonate was added until a pH around 7 was obtained. The
precipitate formed was filtered then purified by flash chromatography
(dichloromethane/methanol: 90/10) to give 6 (75 mg, 21%) as a
beige solid. 1H NMR (400 MHz): 3.39 (m, 4 H), 3.58 (m, 4 H), 4.09
(s, 2 H), 5.21 (s, 1 H), 7.14 (m, 2 H), 7.50 (m, 2 H), 11.86 (br m,
2H). MS: method A; Tr (min) = 0.35; [M + H]+, m/z 312; [M − H]−,
m/z 310.
1
(55 mg, 16%) as an off-white solid. H NMR (400 MHz): 3.31−3.36
(m, 4 H), 3.53−3.58 (m, 4 H), 3.80 (s, 3 H), 4.19 (s, 2 H), 5.21 (s, 1
H), 7.09 (ddd, J = 2.5 and 8.8 and 9.8 Hz, 1 H), 7.37 (dd, J = 2.5 and
9.8 Hz, 1 H), 7.54 (dd, J = 4.8 and 8.8 Hz, 1 H), 11.87 (br m, 1 H).
MS: method A, Tr (min) = 0.44; [M + H]+, m/z 344; [M − H]−, m/z
342 and 12 (68 mg, 20%) as an off-white solid. 1H NMR (400 MHz):
3.27−3.40 (m, partially masked, 4 H), 3.56 (m, 4 H), 3.78 (s, 3 H),
4.18 (s, 2 H), 5.21 (s, 1 H), 7.01 (m, 1 H), 7.44 (dd, J = 2,5 and 9,1
Hz, 1 H), 7.56 (dd, J = 4.9 and 8.8 Hz, 1 H), 11.74 (br m, 1 H). MS:
method A, Tr (min) = 0.46; [M + H]+, m/z 344; [M − H]−, m/z 342.
2-[(5-Bromo-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (13). By a similar procedure to that
described for the synthesis of 6 using 1.3 g (5 mmol) of 2 and 1.87
g (10 mmol) of 4-bromobenzene-1,2-diamine, 13 (0.56 g, 29%) was
obtained as an off-white solid. 1H NMR (400 MHz): 3.34−3.39 (m, 4
H), 3.55−3.59 (m, 4 H), 4.09 (s,2 H), 5.21 (s, 1 H), 7.28 (dd, J = 2.0
and 8.6 Hz, 1 H), 7.47 (d, J = 8.6 Hz, 1 H), 7.71 (d,J = 2.0 Hz, 1 H),
2-(1,3-Benzoxazol-2-ylmethyl)-6-(morpholin-4-yl)pyrimidin-
4(3H)-one (7). Pyridine (4 mL (49.7 mmol)), 550 mg (2.3 mmol) of
EDCI, and 700 mg (6.4 mmol) of 2-aminophenol were added to a
solution of 500 mg (1.9 mmol) of 2 in 4 mL of dimethylformamide.
The reaction mixture was stirred at ambient temperature overnight
and then evaporated to dryness under reduced pressure. Water and
ethyl acetate were added, and the resulting mixture was thus stirred for
30 min. The precipitate formed was filtered off and rinsed with water,
diethyl ether, and petroleum ether. The solid was dried under vacuum
and then suspended in 60 mL of xylene, and 40 mg of 4-
methylbenzenesulphonic acid hydrate were added. The reaction
4799
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
1
12.20 (br m, 2 H). MS: method B, Tr (min) = 2.71; [M + H]+, m/z
390; [M − H]−, m/z 388.
27%) was obtained as an off-white powder. H NMR (400 MHz):
1.35−1.50 (m, 3 H), 1.67 (m, 1 H), 1.79−1.91 (m, 2 H), 2.15−2.30
(m, 2 H), 3.35 (m, 4 H), 3.48 (m, 4 H), 4.45−4.55 (m, 3 H), 5.28 (s, 1
H), 7.42 (m, 2 H), 7.78 (m, 1 H), 8.03 (m, 1 H). MS: method C, Tr
(min) = 0.92; [M + H]+, m/z 394; [M − H]−, m/z 392.
2-[(5-Chloro-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (14). By a similar procedure to that
described for the synthesis of 6 using 400 mg of 2 (1.5 mmol) and
436 mg (3.1 mmol) of 4-chlorobenzene-1,2-diamine, 14 (260 mg,
49%) was obtained as a white powder. 1H NMR (400 MHz): 3.37 (m,
4 H), 3.57 (m, 4 H), 4.10 (s, 2 H), 5.22 (s, 1 H), 7.17 (dd, J = 1.9 and
8.6 Hz, 1 H), 7.51 (d, J = 8.6 Hz, 1 H), 7.57 (d, J = 1.9 Hz, 1 H), 12.18
(br m, 2 H). MS: method A, Tr (min) = 0.51; [M + H]+, m/z 346; [M
− H]−, m/z 344.
6-(Morpholin-4-yl)-2-[(1-phenyl-1H-benzimidazol-2-yl)-
methyl]pyrimidin-4(3H)-one (22). By a similar procedure to that
described for the synthesis of 6 using 300 mg (1.1 mmol) of 2 and 423
mg (2.3 mmol) of N-phenylbenzene-1,2-diamine, 22 (230 mg, 52%)
1
was obtained as a white powder. H NMR (400 MHz): 3.32 (br m, 4
H), 3.56 (m, 4 H), 4.05 (s, 2 H), 5.16 (s, 1 H), 7.11−7.17 (m, 1 H),
7.19−7.30 (m, 2 H), 7.49−7.72 (m, 6 H), 11.74 (br m, 1 H). MS:
method A, Tr (min) = 0.66; [M + H]+, m/z 388; [M − H]−, m/z 386.
2-[(1-Benzyl-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (23). By a similar procedure to that
described for the synthesis of 6 using 300 mg of 2 and 455 mg (2.3
mmol) of N-benzylbenzene-1,2-diamine, 23 (250 mg, 54%) was
2-[(5-Methyl-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (15). By a similar procedure to that
described for the synthesis of 6 using 65 mg (0.2 mmol) of 2 and
30 mg (0.2 mmol) of 3−4-diamino toluene, 15 (7 mg, 8%) was
obtained as an off-white powder. 1H NMR (400 MHz): 2.44 (s, 3 H),
3.20−3.56 (m, 8 H), 4.45 (s, 2 H), 5.30 (s, 1 H), 7.25−7.70 (m, 3 H).
MS: method C, Tr (min) = 0.74; [M + H]+, m/z 326; [M − H]−, m/z
324.
2-[(5-Methoxy-1H-benzimidazol-2-yl)methyl]-6-(morpholin-
4-yl)pyrimidin-4(3H)-one (16). By a similar procedure to that
described for the synthesis of 6 using 261 mg (1 mmol) of 2 and 166
mg (1.2 mmol) of 3,4-diaminoanisole, 16 (8 mg, 2%) was obtained as
an off-white powder (purity = 90%). 1H NMR (400 MHz): 3.38 (m, 4
H), 3.58 (m, 4 H), 3.76 (s, 3 H), 4.04 (s,2 H), 5.21 (s, 1 H), 6.72−
6.82 (m, 1 H), 6.89−7.15 (br m, 1 H), 7.27−7.52 (br m, 1 H), 11.82
(br m, 1 H), 12.11 (br m, 1 H). MS: method A, Tr (min) = 0.40; [M +
H]+, m/z 342.
1
obtained as a beige powder. H NMR (400 MHz): 3.25−3.35 (m
partially masked, 4 H), 3.52−3.58 (m, 4 H), 4.18 (s, 2 H), 5.18 (s, 1
H), 5.56 (s, 2 H), 7.13−7.20 (m, 4 H), 7.23−7.35 (m, 3 H), 7.39−7.47
(m, 1 H), 7.55−7.64 (m, 1 H), 11.88 (br m, 1), MS: method A, Tr
(min) = 0.58; [M + H]+, m/z 402; [M − H]−, m/z 400.
6-(Morpholin-4-yl)-2-[(4-phenyl-1H-benzimidazol-2-yl)-
methyl]pyrimidin-4(3H)-one (24). By a similar procedure to that
described for the synthesis of 6 using 840 mg (3.2 mmol) of 2 and 730
mg (2.9 mmol) of 1,1′-biphenyl-2,3-diamine, 24 (455 mg, 40%) was
obtained as a gray powder. 1H NMR (400 MHz): 3.22−3.42 (m
partially masked, 4 H), 3.56 (m, 4 H), 4.14 (s, 2 H), 5.20 (s, 1 H),
6.93−8.26 (m, 8 H), 11.80 (br m, 1 H), 12.30 (br m, 1 H). MS:
method A, Tr (min) = 0.61; [M + H]+, m/z 388; [M − H]−, m/z 386.
2-[(1-Methyl-4-phenyl-1H-benzimidazol-2-yl)methyl]-6-
(morpholin-4-yl)pyrimidin-4(3H)-one (25) and 2-[(1-Methyl-7-
phenyl-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-yl)-
pyrimidin-4(3H)-one (26). By a similar procedure to that described
for the synthesis of 11 and 12 using 300 mg (0.8 mmol) of 24, two
compounds were obtained. 25 (57 mg, 20%) as an off-white solid. 1H
NMR (400 MHz): 3.30 (m partially masked, 4 H), 3.54 (m, 4 H), 3.86
(s, 3 H), 4.28 (s, 2 H), 5.19 (s, 1 H), 7.25−7.62 (m, 6H), 8.04 (d, J =
7.8 Hz, 2 H), 11.93 (br m, 1 H). MS: method A, Tr (min) = 0.62; [M
+ H]+, m/z 402; [M − H]−, m/z 400. 26 (52 mg, 18%) as a white
6-(Morpholin-4-yl)-2-{[5-(trifluoromethyl)-1H-benzimidazol-
2-yl]methyl}pyrimidin-4(3H)-one (17). By a similar procedure to
that described for the synthesis of 6 using 65 mg (0,2 mmol) of 2 and
44 mg (0,2 mmol) of 4-(trifluoromethyl)-1,2-phenylenediamine, 17
1
(15 mg, 16%) was obtained as an off-white powder. H NMR (400
MHz): 3.30 (m, 4 H), 3.53 (m, 4 H), 4.20 (s, 2 H), 5.25 (s, 1 H), 7.50
(m, 1 H), 7.75 (m, 1H), 7.92 (s, 1H). MS: method C, Tr (min) = 0.93;
[M + H]+, m/z 380; [M − H]−, m/z 378.
2-[(1-Ethyl-1H-benzimidazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (18). By a similar procedure to that
described for the synthesis of 11 or 12 using 155 mg (0.5 mmol) of
6, 0.38 mL (0.76 mmol) of sodium hydroxyde (2N) and 60 μL (1
mmol) of iodoethane, 18 (30 mg, 18%) was obtained as a beige
1
solid. H NMR (400 MHz): 3.32 (m, 7 H), 3.57 (m, 4 H), 4.31 (s, 2
1
H), 5.24 (s, 1 H), 7.16 (d, J = 7.8 Hz, 1 H), 7.36 (t, J = 7.8 Hz, 1 H),
7.42 (m, 2 H), 7.46−7.56 (m, 3 H), 7.70 (d, J = 7.8 Hz, 1 H), 11.96
(br m, 1 H). MS: method A, Tr (min) = 0.68; [M + H]+, m/z 402; [M
− H]−, m/z 400.
2-[(1-Methyl-5-phenyl-1H-benzimidazol-2-yl)methyl]-6-
(morpholin-4-yl)pyrimidin-4(3H)-one (27). By a similar procedure
to that described for the synthesis of 6 using 733 mg (2.8 mmol) of 2
and 455 mg (2.3 mmol) of N-4-methyl-biphenyl-3,4-diamine, 27 (674
mg, 66%) was obtained as a white powder. 1H NMR (400 MHz): 3.35
(m, 4 H), 3.57 (m, 4 H), 3.83 (s, 3 H), 4.22 (s, 2 H), 5.22 (s, 1 H),
7.32 (t, J = 7.8 Hz, 1 H), 7.45 (t, J = 7.8 Hz, 2 H), 7.55 (d, J = 8.5 Hz,
1 H), 7.61 (d, J = 8.5 Hz, 1 H), 7.69 (d, J = 7.8 Hz, 2 H), 7.84 (s, 1 H),
11.89 (br m, 1 H). MS: method A, Tr (min) = 0.63; [M + H]+, m/z
402; [M − H]−, m/z 400.
2-[(1-Methyl-6-phenyl-1H-benzimidazol-2-yl)methyl]-6-
(morpholin-4-yl)pyrimidin-4(3H)-one (28). By a similar procedure
to that described for the synthesis of 6 using 710 mg (2.7 mmol) of 2
and 490 mg of N-3-methyl-biphenyl-2,3-diamine, 28 (440 mg, 44%)
was obtained as an off-white powder. 1H NMR (400 MHz): 3.30−3.40
(m partially masked, 4 H), 3.55 (m, 4 H), 3.92 (s, 3 H), 4.34 (s, 2 H),
5.25 (s, 1 H), 7.37 (t, J = 7.8 Hz, 1 H), 7.49 (t, J = 7.8 Hz, 2 H), 7.61
(d, J = 8.3 Hz, 1 H), 7.70 (br d, J = 8.3 Hz, 1 H), 7.77 (d, J = 7.8 Hz, 2
H), 7.95 (s large, 1 H); 11.95 (br m, 1 H). MS: method A, Tr (min) =
0.67; [M + H]+, m/z 402; [M − H]−, m/z 400.
powder. H NMR (400 MHz): 1.31 (t, J = 7.1 Hz, 3 H), 3.34 (m, 4
H), 3.57 (m, 4 H), 4.18 (s, 2 H), 4.31 (q, J = 7.1 Hz, 2 H), 5.21 (s, 1
H), 7.13−7.19 (m, 1 H), 7.20−7.25 (m, 1 H), 7.51−7.59 (m, 2 H),
11.88 (br m, 1 H). MS: method A, Tr (min) = 0.42; [M + H]+, m/z
340; [M − H]−, m/z 338.
2-[(1-Cyclopropyl-5-fluoro-1H-benzimidazol-2-yl)methyl]-6-
(morpholin-4-yl)pyrimidin-4(3H)-one (19). By a similar procedure
to that described for the synthesis of 6 using 300 mg (1.1 mmol) of 2
and 645 mg (3.9 mmol) of N1-cyclopropyl-4-fluorobenzene-1,2-
1
diamine, 19 (15 mg, 4%) was obtained as a white powder. H NMR
(400 MHz): 1.04−1.22 (m, 4 H), 3.33−3.43 (m, 5 H), 3.54−3.60 (m,
4 H), 4.24 (s, 2 H), 5.22 (s, 1 H), 7.09 (dt, J = 2.4 and 9.3 Hz, 1 H),
7.39 (dd, J = 2.4 and 9.8 Hz, 1 H), 7.56 (dd, J = 4.8 and 9.3 Hz, 1 H),
11.80 (br m, 1 H). MS: method A, Tr (min) = 0.57; [M + H]+, m/z
370; [M − H]−, m/z 368.
2-{[5-Fluoro-1-(2-methoxyethyl)-1H-benzimidazol-2-yl]-
methyl}-6-(morpholin-4-yl)pyrimidin-4(3H)-one (20). By a sim-
ilar procedure to that described for the synthesis of 6 using 300 mg
(1.1 mmol) of 2 and 761 mg (4.1 mmol) of 4-fluoro-N1-(2-
methoxyethyl)benzene-1,2-diamine hydrochloride, 20 (120 mg, 27%)
was obtained as a white powder. 1H NMR (400 MHz): 3.20 (s, 3 H),
3.33−3.40 (m, 4 H), 3.54−3.59 (m, 4 H), 3.63 (t, J = 5.3 Hz, 2 H),
4.19 (s, 2 H), 4.47 (t, J = 5.3 Hz, 2 H), 5.21 (s, 1 H), 7.08 (dt, J = 2.4
and 9.3 Hz, 1 H), 7.37 (dd, J = 2.4 and 9.8 Hz, 1 H), 7.57 (dd, J = 4.8
and 9.3 Hz, 1 H), 11.80 (br m, 1 H). MS: method A, Tr (min) = 0.51;
[M + H]+, m/z 388; [M − H]−, m/z 386.
2-[(4-Fluoro-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (29). By a similar procedure to that
described for the synthesis of 7 using 500 mg (1.9 mmol) of 2 and
357 mg (2.8 mmol) of 2-amino-3-fluorophenol, 32 (110 mg, 17%) was
obtained as a white solid. H NMR (400 MHz): 3.35 (m, 4 H), 3.56
(m, 4 H), 4.29 (s, 2 H), 5.26 (s, 1 H), 7.25 (dd, J = 8.3 and 10.2 Hz, 1
2-[(1-Cyclohexyl-1H-benzimidazol-2-yl)methyl]-6-(morpho-
lin-4-yl)pyrimidin-4(3H)-one (21). By a similar procedure to that
described for the synthesis of 6 using 65 mg (0.2 mmol) of 2 and 47
mg (0.2 mmol) of N1-cyclohexyl-1,2-benzenediamine, 21 (27 mg,
1
4800
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
H), 7.42 (dt, J = 5.1 and 8.3 Hz, 1 H), 7.60 (d, J = 8.3 Hz, 1 H), 11.93
(br s, 1 H). MS: method A, Tr (min) = 0.64; [M + H]+, m/z 331; [M
− H]−, m/z 329.
H). MS: method A, Tr (min) = 0.69; [M + H]+, m/z 327; [M − H]−,
m/z 325.
2-[(4-Bromo-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (38). By a similar procedure to that
described for the synthesis of 7 using 5.0 g (19.1 mmol) of 2 and
3.6 g (19.1 mmol) of 2-amino-3-bromophenol, 38 (1.22 g, 16%) was
obtained as a white solid. 1H NMR (400 MHz): 2.50 (masked s, 3 H),
3.35 (m, 4 H), 3.56 (m, 4 H), 4.31 (s, 2 H), 5.26 (br s, 1 H), 7.35 (t, J
= 8.1 Hz, 1 H), 7.61 (d, J = 8.1 Hz, 1 H), 7.75 (d, J = 8.1 Hz, 1 H),
11.94 (br m, 1 H). MS: method A, Tr (min) = 0.70; [M + H]+, m/z
393; [M − H]−, m/z 391.
2-[(5-Fluoro-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (30). By a similar procedure to that
described for the synthesis of 7 using 1 g (3.8 mmol) of 2 and 466
mg (3.7 mmol) of 2-amino-4-fluorophenol, 30 (188 mg, 15%) was
1
obtained as a brown solid. H NMR (400 MHz): for this product, all
the signals were broad, with: 3.33−3.39 (m, 4 H), 3.54−3.62 (m, 4 H),
4.26 (s, 2 H), 5.25 (s, 1 H), 7.25 (t, J = 9.0 Hz, 1 H), 7.61 (d, J = 9.0
Hz, 1 H), 7.76 (dd, J = 4,5 and 9.0 Hz, 1 H), 11.83−11.96 (br m, 1 H).
MS: method A, Tr (min) = 0.65; [M + H]+, m/z 331; [M − H]−, m/z
329.
6-(Morpholin-4-yl)-2-[(4-phenyl-1,3-benzoxazol-2-yl)-
methyl]pyrimidin-4(3H)-one (39). A mixture of 38 (130 mg, 0.332
mmol), phenylboronic acid (45 mg, 0.4 mmol), Pd(PPh3)4 (38 mg,
0.03 mmol), sodium carbonate (83 mg, 0.8 mmol) in water (3 mL),
and 1,2-dimethoxyethane (5 mL) was stirred at 100 °C for 5 h and
then evaporated to dryness under reduced pressure. Water and
dichloromethane were added to the residue, and the organic layer was
washed with brine and dried over anhydrous magnesium sulfate. The
filtrate was concentrated, and the residue was purified by flash
chromatography (dichloromethane/methanol: 95/05) to give 39 (68
2-[(6-Fluoro-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (31). By a similar procedure to that
described for the synthesis of 7 using 550 mg (2.1 mmol) of 2 and
815 mg (6.4 mmol) of 2-amino-5-fluorophenol, 31 (15 mg, 2%) was
1
obtained as a white solid.; H NMR (400 MHz): 3.35 (m, 4 H), 3.56
(m, 4 H), 4.25 (s, 2 H), 5.25 (br s, 1 H), 7.24 (ddd, J = 2.6 and 8.7 and
10.0 Hz, 1 H), 7.70−7.77 (m, 2 H), 11.93 (br s, 1 H). MS: method A,
Tr (min) = 0.65; [M + H]+, m/z 329; [M − H]−, m/z 331.
2-[(7-Fluoro-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (32). By a similar procedure to that
described for the synthesis of 7 using 500 mg (1.9 mmol) of 2 and
357 mg (2.8 mmol) of 2-amino-6-fluorophenol, 32 (110 mg, 17%) was
obtained a white powder. 1H NMR (400 MHz): 3.34−3.38 (m, 4 H),
3.53−,3.59 (m, 4 H), 4.31 (s, 2 H), 5.26 (s, 1 H), 7.30−7.42 (m, 2 H),
7.59 (dd, J = 1.4 and 7.7 Hz, 1 H), 11.93 (br m, 1 H). MS: method B,
Tr (min) = 3.11; [M + H]+, m/z 331; [M − H]−, m/z 329.
2-(1,3-Benzoxazol-2-ylmethyl)-3-methyl-6-(morpholin-4-yl)-
pyrimidin-4(3H)-one (33). By a similar procedure to that described
for the synthesis of 7 using 261 mg (0.9 mmol) of 4 and 350 mg (3.2
mmol) of 2-aminophenol, 33 (40 mg, 12%) was obtained as a pale-
yellow powder, mp 191 °C. 1H NMR (400 MHz): 3.26 (m, 4 H), 3.40
(s, 3 H), 3.50 (m, 4 H), 4.60 (s, 2 H), 5.37 (s, 1 H), 7.38 (m, 2 H),
7.72 (m, 2 H). MS: method B, Tr (min) = 3.11; [M + H]+, m/z 327;
[M − H]−, m/z 325.
2-[(4-Methyl-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (34). By a similar procedure to that
described for the synthesis of 7 using 1 g (3.8 mmol) of 2 and 471
mg (3.8 mmol) of 2-amino-3-methylphenol, 34 (286 mg, 23%) was
obtained as a beige solid. 1H NMR (400 MHz): 2.50 (masked s, 3 H),
3.24−3.39 (m partially masked, 4 H), 3.56 (m, 4 H), 4.25 (s, 2 H),
5.25 (br s, 1 H), 7.17 (d, J = 7.8 Hz, 1 H), 7.26 (t, J = 7.8 Hz, 1 H),
7.49 (d, J = 7.8 Hz, 1 H), 11.80 (br m, 1 H). MS: method A, Tr (min)
= 0.72; [M + H]+, m/z 327; [M − H]−, m/z 325.
1
mg, 53%) as a white solid. H NMR: 3.34 (m, 4 H), 3.55 (m, 4 H),
4.32 (s, 2 H), 5.24 (s, 1 H), 7.41 (t, J = 7.8 Hz, 1 H), 7.45−7.54 (m, 3
H), 7.61 (d, J = 7.8 Hz, 1 H), 7.70 (d, J = 7.8 Hz, 1 H), 7.98 (d, J = 7.8
Hz, 2 H), 11.92 (br m, 1 H). MS: method A, Tr (min) = 0.88; [M +
H]+, m/z 389; [M − H]−, m/z 387.
6-(Morpholin-4-yl)-2-[(7-phenyl-1,3-benzoxazol-2-yl)-
methyl]pyrimidin-4(3H)-one (40). By a similar procedure to that
described for the synthesis of 7 using 330 mg (1.3 mmol) of 2 and 467
mg (2.5 mmol) of 2-amino-6-phenylphenol, 40 (73 mg, 15%) was
1
obtained as a pale-green solid. H NMR (400 MHz): 3.36 (m, 4 H),
3.54 (m, 4 H), 4.30 (s, 2 H), 5.24 (br s, 1 H), 7.41−7.50 (m, 2 H),
7.55 (t, J = 7.8 Hz, 2 H), 7.65 (d, J = 7.8 Hz, 1 H), 7.71 (d, J = 7.8 Hz,
1 H), 7.88 (d, J = 7.8 Hz, 2 H), 11.91 (br m, 1 H). MS: method A, Tr
(min) = 0.85; [M + H]+, m/z 389; [M − H]−, m/z 387.
6-(Morpholin-4-yl)-2-{[4-(thiophen-2-yl)-1,3-benzoxazol-2-
yl]methyl}pyrimidin-4(3H)-one (41). By a similar procedure to that
described for the synthesis of 39 using 200 mg (0.51 mmol) of 38 and
73 mg (0.57 mmol) of thiophene-2-boronic acid, 41 (30 mg, 15%) was
obtained as a gray solid. 1H NMR (400 MHz): 3.20−3.40 (m partially
masked, 4 H), 3.55 (m, 4 H), 4.35 (s, 2 H), 5.24 (s, 1 H), 7.21 (dd, J =
3.7 and 5.1 Hz, 1 H); 7.42 (t, J = 7.8 Hz, 1 H), 7.60−7.71 (m, 3 H),
8.00 (dd, J = 1.2 and 3.7 Hz, 1 H), 12.01 (br m, 1 H). MS: method A,
Tr (min) = 0.87; [M + H]+, m/z 395; [M − H]−, m/z 393.
6-(Morpholin-4-yl)-2-{[4-(pyridin-2-yl)-1,3-benzoxazol-2-yl]-
methyl}pyrimidin-4(3H)-one (42). By a similar procedure to that
described for the synthesis of 7 using 550 mg (2.1 mmol) of 2 and 460
mg (2.5 mmol) of 2-amino-3-(pyridin-2-yl)phenol (102 mg, 12%), 42
2-[(5-Methyl-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (35). By a similar procedure to that
described for the synthesis of 7 using 500 mg (1.9 mmol) of 2 and
349 mg (2.8 mmol) of 2-amino-4-methylphenol, 35 (280 mg, 45%)
was obtained as a white solid, mp 242 °C. 1H NMR (400 MHz): 2.41
(s, 3 H), 3.35 (m, 4 H), 3.56 (m, 4 H), 4.22 (s, 2 H), 5.24 (br s, 1 H),
7.19 (dd, J = 1.9 and 8.3 Hz, 1 H), 7.51 (d, J = 1.9 Hz, 1 H), 7.57 (d, J
= 8.3 Hz, 1 H), 11.91 (br m, 1 H). MS: method A, Tr (min) = 0.74;
[M + H]+, m/z 327; [M − H]−, m/z 325.
1
was obtained as pale yellow solid. H NMR (400 MHz): 3.20−3.40
(m, partially masked, 4 H), 3.54 (m, 4 H), 4.37 (s, 2 H), 5.26 (s, 1 H),
7.40 (m, 1 H), 7.52 (t, J = 7.8 Hz, 1 H), 7.80 (dd, J = 1.0 and 7.8 Hz, 1
H), 7.96 (dt, J = 2.0 and 7.8 Hz, 1 H), 8.24 (dd, J = 1.0 and 7.8 Hz, 1
H), 8.69 (d, J = 7.8 Hz, 1 H), 8.74 (m, 1 H), 11.92 (br m, 1 H). MS:
method A, Tr (min) = 0.52; [M + H]+, m/z 390; [M − H]−, m/z 388.
6-(Morpholin-4-yl)-2-{[4-(pyridin-3-yl)-1,3-benzoxazol-2-yl]-
methyl}pyrimidin-4(3H)-one (43). By a similar procedure to that
described for the synthesis of 39 using 130 mg (0.33 mmol) of 38 and
45 mg (0.37 mmol) of pyridine-3-boronic acid, 43 (30 mg, 23%) was
2-[(6-Methyl-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (36). By a similar procedure to that
described for the synthesis of 7 using 400 mg (1.5 mmol) of 2 and
466 mg (3.8 mmol) of 2-amino-5-methylphenol, 36 (105 mg, 21%)
1
1
obtained as a white solid. H NMR (400 MHz): 3.35 (m, 4 H), 3.55
was obtained as a white solid. H NMR (400 MHz): 2.43 (s, 3 H),
(m, 4 H), 4.33 (m, 2 H), 5.25 (s, 1 H), 7.50−7.58 (m, 2 H), 7.70 (d, J
= 7.8 Hz, 1 H), 7.78 (d, J = 8.1 Hz, 1 H), 8.36 (td, J = 2.0 and 7.8 Hz,
1 H), 8.61 (dd, J = 1.7 and 4.9 Hz, 1 H), 9.19 (d, J = 2.0 Hz, 1 H),
11.94 (br m, 1 H). MS: method A, Tr (min) = 0.49; [M + H]+, m/z
390; [M − H]−, m/z 388.
3.36 (m, 4 H), 3.56 (m, 4 H), 4.21 (s, 2 H), 5.24 (br m, 1 H), 7.18 (br
d, J = 8.3 Hz, 1 H), 7.51 (br s, 1 H), 7.57 (d, J = 8.3 Hz, 1 H), 11.91
(br m, 1 H). MS: method A, Tr (min) = 0.69; [M + H]+, m/z 327; [M
− H]−, m/z 325.
2-[(7-Methyl-1,3-benzoxazol-2-yl)methyl]-6-(morpholin-4-
yl)pyrimidin-4(3H)-one (37). By a similar procedure to that
described for the synthesis of 7 using 1 g (3.8 mmol) of 2 and 680
mg (5.5 mmol) of 2-amino-6-methylphenol, 37 (835 mg, 67%) was
obtained as a beige solid. 1H NMR (400 MHz): 2.47 (s, 3 H), 3.37 (m,
4 H), 3.57 (m, 4 H), 4.25 (s, 2 H), 5.25 (s, 1 H), 7.20 (d, J = 7.8 Hz, 1
H), 7.25 (t, J = 7.8 Hz, 1 H), 7.52 (d, J = 7.8 Hz, 1 H), 11.90 (br s, 1
6-(Morpholin-4-yl)-2-{[4-(pyridin-4-yl)-1,3-benzoxazol-2-yl]-
methyl}pyrimidin-4(3H)-one (44). By a similar procedure to that
described for the synthesis of 39 using 200 mg (0.51 mmol) of 38 and
115 mg (0.94 mmol) of pyridine-4-boronic acid pinacol ester, 44 (30
1
mg, 15%) was obtained as a beige solid. H NMR (400 MHz): 3.20−
3.40 (m, partially masked, 4 H), 3.54 (m, 4 H), 4.35 (s, 2 H), 5.25 (s, 1
4801
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
H), 7.54 (t, J = 7.8 Hz, 1 H), 7.79 (d, J = 7.8 Hz, 1 H), 7.83 (d, J = 7.8
Hz, 1 H), 8.04 (d, J = 5.9 Hz, 2 H), 8.70 (d, J = 5.9 Hz, 2 H), 11.95 (br
m, 1 H). MS: method A, Tr (min) = 0.45; [M + H]+, m/z 390; [M −
H]−, m/z 388.
PI(4,5)P2 substrate/enzyme mixture before starting the reaction by
the addition of 100 μM ATP. Enzyme concentrations corresponded to
100 pM, 15 pM, 30 pM, and 400 pM for α, β, δ, and γ isoforms,
respectively. After 15 min of incubation at RT, the reaction was
stopped by adding a mixture of PIP3-biotin and stop solution
(Millipore Kit). Detection mixture, containing streptavidin-coupled
allophycocyanin (SA-XL665) and precombined GST-pleckstrin
homology (GST-PH) domain with Anti-GST-europium labeled, was
then added. After overnight incubation at 4 °C, HTRF signals were
measured with a Rubystar (BMG Labtech). For this competition
binding assay, wells without enzyme were used to calculate the
maximal signal (max). The nontreated wells containing 3% DMSO
were used to determine the minimum signal (min). The percentage
inhibition was calculated for each concentration of compound
according to the following formula [% inhibition = ((Test − min)/
(max − min)) × 100]. The activity of the product was estimated by
using the concentration of compound where percent inhibition is
equal to 50 (IC50) obtained from a dose−response curve with 11
increasing concentrations tested in single replicates and fitted with
XLFit 4 software from Excel version 4.2.2 using the 4-parameter
logistic model (eq 205). IC50 values represent the mean from at least
two independent experiments; when more than a 3-fold difference was
observed between the two independent values, additional experiments
were performed in order to determine the mean IC50.
2-[(5-Fluoro-4-phenyl-1,3-benzoxazol-2-yl)methyl]-6-(mor-
pholin-4-yl)pyrimidin-4(3H)-one (45). By a similar procedure to
that described for the synthesis of 7 using 630 mg (2.4 mmol) of 2 and
445 mg (2.2 mmol) of 2-amino-6-fluorobiphenyl-3-ol (see Scheme 6),
1
(127 mg, 13%), 45 was obtained as a beige solid. H NMR (400
MHz): 3.34 (m, 4 H), 3.55 (m, 4 H), 4.29 (s, 2 H), 5.23 (br s, 1 H),
7.38 (dd, J = 9.0 and 11.2 Hz, 1 H), 7.45 (t, J = 7.8 Hz, 1 H), 7.52 (t, J
= 7.8 Hz, 2 H), 7.68 (d, J = 7.8 Hz, 2 H), 7.76 (dd, J = 4.0 and 9.0 Hz,
1 H), 11.94 (br m, 1 H). MS: method A, Tr (min) = 0.88; [M + H]+,
m/z 407; [M − H]−, m/z 405.
2-Amino-6-fluorobiphenyl-3-ol. To a suspension of 16 g (84
mmol) of 3-bromo-4-fluorophenol and 16 g of ammonium nickel-
(II)sulfate hexahydrate (41 mmol) in 100 mL of dichloromethane was
added 8.1 g of nitric acid (d = 1.41, 89.6 mmol) over 8 min while
maintaining the internal temperature below 25 °C with an ice bath.
The resulting mixture was allowed to stir for 20 min and poured into
250 g of crushed ice. The layers are separated and the aqueous phase
extracted with dichloromethane. The organic layers were combined,
washed with brine, dried over anhydrous magnesium sulfate, and then
concentrated under reduced pressure. The residue was purified by
flash chromatography; eluent, dichloromethane/heptane 70/30 to give
3-bromo-4-fluoro-2-nitrophenol (6.1 g, 31%) as a yellow solid. 1H
NMR (300 MHz): 7.11 (dd, J = 4.3 and 9.3 Hz, 1 H), 7.46 (dd, J = 8.5
and 9.3 Hz, 1 H), 11.41 (br s, 1 H).
Akt Phosphorylation Cell-Based Assays. Cell lines were
purchased from ATCC.36 Cells were cultured in DMEM medium,
containing 10% fetal calf serum and 1% glutamine (complete culture
medium).
A mixture of 5.2 g (22 mmol) of 3-bromo-4-fluoro-2-nitrophenol,
5.5 g (27 mmol) of 4,4,5,5-tetramethyl-2-phenyl-1,3,2-dioxaborolane,
1,16 g (1.6 mmol) of Pd(dppf)Cl2, 21.5 g (66 mmol) of cesium
carbonate, 43 mL of water, and 166 mL of dioxane was stirred at 95 °C
for 4 h. After cooling to room temperature, water, and ethyl were
added and the insoluble material formed filtered off. The organic layer
was washed with brine and dried over anhydrous magnesium sulfate.
The filtrate was concentrated under reduced pressure, and the residue
was purified by flash chromatography; eluent, dichloromethane/
heptane 50/50 then 100% dichloromethane to give 6-fluoro-2-
nitrobiphenyl-3-ol (4.5 g, 88%) as a yellow solid. 1H NMR (300
MHz): 7.10 (dd, J = 4.4 and 9.2 Hz, 1 H); 7.29−7.35 (m, 2 H), 7.41
(dd, J = 9.1 and 9.2 Hz, 1 H), 7.43−7.50 (m, 3 H), 11.04 (br s, 1 H).
To a solution of 4.4 g (19 mmol) of 6-fluoro-2-nitrobiphenyl-3-ol in
75 mL of acetic acid, 5.1 g (91 mmol) of iron powder were added. The
reaction mixture was stirred at 70 °C during 30 min and then
evaporated to dryness under reduced pressure. The residue was taken
up in a mixture of water and ethyl acetate. The insoluble material was
filtered off and the aqueous layer extracted with ethyl acetate. The
organic layers were combined, washed with brine, dried over
anhydrous magnesium sulfate, and then concentrated under reduced
pressure. The residue was purified by flash chromatography; eluent,
heptane/ethyl acetate 80/20 to give 2-amino-6-fluorobiphenyl-3-ol
(3.0 g, 78%) as a yellow solid. 1H NMR (300 MHz): 4.19 (br s, 2 H),
6.31 (dd, J = 8.6 and 9.5 Hz, 1 H), 6.63 (dd, J = 5.4 and 8.6 Hz, 1 H),
7.29−7.50 (m, 5 H), 9.19 (br s, 1 H).
In Vitro PI3K Enzyme Assay. Human p110α and δ with N-
terminal poly-His tags were coexpressed with p85α in a Sf9
baculovirus expression system, and the p110α, δ/p85α heterodimers
were purified by sequential Ni-NTA and heparin chromatography.
Human p110β with N-terminal poly-His tag was coexpressed with
p85α in a S21 baculovirus expression system, and the p110β/p85α
heterodimer was purified by sequential Ni-NTA, A-EX (Q-HP), and
Superdex S200 chromatography. A truncated form of human PI3Kγ
encompassing residues 114−1102, N-terminally labeled with poly-His
tag, was expressed with baculovirus in Sf9 insect cells and purified by
sequential Ni-NTA, Superdex-200 chromatography. Lipid kinase
activity assays were performed using PI3K HTRF (homogeneous
time-resolved fluorescence) Millipore Kit in 384-well format according
to manufacturer instructions. The assay was performed in 10 μL
reaction volume: serial dilutions of inhibitors (3% DMSO) were
preincubated for 15 min at room temperature (RT) with 10 μM
MEF-3T3-myristoylated p110β cell line was an inducible stable
clone established by transfecting a pTRE2Hygro-myc vector and
human myr wt PI3KCb sequence of interest in Mouse Embryonnary
Fibroblast MEF3T3 Tet Off (Clontech 631139). Cells were cultured
in DMEM medium supplemented with 10% tet system approved FCS,
2 mM L-glutamine, 100 μg/mL geneticin, 200 μg/mL hygromycin B.
MEF-3T3-myristoylated p110δ cell line was an inducible stable clone
established by transfecting a pTRE2Hygro-myc vector and human myr
wt PI3KCd sequence of interest in Mouse Embryonnary Fibroblast
MEF3T3 Tet Off (Clontech 631139). Cells were cultured in the same
medium as described for MEF-3T3-myristoylated p110β. Akt
phosphorylation test was based on a sandwich immunoassay using
the MSD Multispot Biomarker Detection kit from Meso Scale
Discovery “phospho-Akt (Ser473) whole cell lysate kit” (no.
K151CAD).The primary antibody specific for pAkt-S473 was coated
onto an electrode in each well of the 96-well plates of the MSD kit.
After the addition of a protein lysate to each well, the signal was
visualized by adding a secondary detection antibody labeled with an
electrochemiluminescent compound. The procedure followed was that
described in the MSD kit. On day 1, cells were seeded into 96-well
plates at the concentration of 5000−35000 cells/well depending on
the cell line (PC3 and H460 cells), 20000 cells/well (MEF-3T3-
myristoylated p110δ and MEF-3T3-myristoylated p110β cells) or
320000 cells/well (Raw 264.7 cells) in 200 μL of DMEM complete
medium (PC3, MEF-3T3-myristoylated p110β and MEF-3T3-
myristoylated p110δ cells) or in serum-free medium (Raw 264−7
cells), and incubated at 37 °C, 5% CO2, overnight. On day 2, cells
were incubated in the presence or absence of the test products for 30
min to 2 h at 37 °C in the presence of 5% of CO2. In the case of Raw
264.7, at the end of compound treatment, cells were stimulated for 5
min with C5a at 10−8 M final concentration, at 37 °C in the presence
of 5% of CO2. In the case of MEF-3T3-myristoylated p110δ, cells were
serum-deprived for 6 h before treatment with test compounds in
serum-free medium. The test compounds, diluted in DMSO, were
added from a 20-time concentrated stock solution, the final DMSO
concentration being 0.1%. The compounds were tested at increasing
concentrations ranging from less than 1 nM to 10 μM. After
incubation, cells were lysed for the preparation of the proteins,
according to the MSD kit description. The plates were read on the
S12400 instrument from Meso Scale Discovery. Wells without cells
and containing the lysis buffer were used to determine the background
noise subtracted from all the measurements (min). The wells
4802
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
containing cells in the absence of product and in the presence of 0.1%
DMSO were used to determine the 100% signal (max). The
percentage inhibition was calculated for each concentration of
compound according to the following formula [% inhibition = (1 −
(Test − min)/(max − min)) × 100]. The activity of the product was
estimated by using the IC50 obtained from a dose−response curve with
10 concentrations tested in monoplicate and fitted with XLFit 4
software from Excel version 4.2.2 using the 4-parameter logistic model
(eq 205). IC50 values represent the mean from at least two
independent experiments; when more than a 3-fold difference was
observed between the two independent values, additional experiments
were performed in order to determine the mean IC50.
027), 100 mM NaCl (Sigma S5150), 1 mM EDTA pH 8 (InVitrogen
15575−038), 1 mM EGTA pH 8 (Sigma E4378), 1% Triton (Sigma
T9284), 1 mM NaF (Sigma S7920), 20 mM Na4P2O7 (Sigma
S6422), 1 mM activated Na3VO4 (Sigma S6508), 10% glycerol
(Fisher Scientific G/0650/17) supplemented with protease inhibitor
mix (Roche 11836145001)] by mechanical dissociation using the
Precellys 24 bead beating homogenizer (Ozyme, BER1011) set at two
runs of 2 × 30 s at 6000 rpm at 4 °C and connected to a Cryolys
cooling system. The extracts were incubated 2 h at 4 °C to extract
membrane proteins, clarified by a 16000g centrifugation for 15 min at
4 °C, and supernatants were collected, aliquoted, snap-frozen in dry
ice, and stored at −80 °C until analysis.
Protein concentrations were determined by the Bradford method
(Kit Pierce BCA 23225) according to the manufacturer instructions.
Tumors extract samples were not diluted for the assays. Levels of total
protein Akt and phospho protein pAkt-S473 and pAkt-T308 were
determined using Mesoscale electrochemiluminescence duplex
immunoassays according to manufacturer instructions. The impact
on biomarker upon compound treatment was expressed as the mean
modulation of target kinase phosphorylation in treated tumors as
compared to the control (animals treated with vehicle) and was
calculated as [100 − ((mean phospho protein for treated group)/
(mean phospho protein for vehicle group)) × 100]. Results of
percentages of inhibition are reported as the mean and standard error
of the mean of three mice per group.
Xenograft Studies. Human prostate PC-3 cells were purchased at
ATCC (Batch CTRL-1435, Rockville, MD, USA). The PC-3 was
initiated from a bone metastasis of grade V prostatic adenocarcinoma
from a 62-year-old male Caucasian. The PC-3 cells were cultured in F-
12 medium + 10% fetal calf serum (FCS) + 2 mL glutamin. The tumor
model was established by implanting subcutaneously (sc) 3 × 106 cells
mixed with 50% matrigel (Reference 356234, Becton Dickinson
Biosciences) per SCID female mice and was maintained by sc serial
passages once every 3 weeks in SCID female mice. Animals bearing
tumors were randomized according to a given median tumor burden
range on a specific day post tumor implantation, and the compound
was administered orally at 300 mg/kg twice a day (bid) in suspension
prepared in 0.6% methylcellulose (MC) and 0.5% Tween 80, with 10
animals per group. Tumor volume was measured regularly during the
treatment period.
Changes in tumor volume for each treated (T) and control (C)
group were calculated for each tumor by subtracting the tumor volume
on the day of first treatment (staging day) from the tumor volume on
the specified observation day. The median ΔT was calculated for the
treated group, and the median ΔC was calculated for the control
group. Then the ratio ΔT/ΔC was calculated and expressed as a
percentage.
The measurement of Akt phosphorylation on threonine 308 residue
(pAkt-T308) was evaluated by Western blotting using a primary
antibody specific for pAkt-T308. Briefly, on day 1, PC3 cells were
seeded into 6-well microplates at 800000 cells per well, in complete
culture medium containing 10% FCS and incubated at 37 °C, 5% CO2,
overnight. On day 2, the cells were incubated in the presence or
absence of compound 8 for 30 min to 2 h at 37 °C in the presence of
5% of CO2. At the end of cell treatment period, adherent cells were
lysed for the preparation of the proteins. Cells were lysed in a lysis
buffer containing Hepes 50 mM, NaCl 150 mM, glycerol 10%, Triton
X100 1%, pH = 7.5, adding extemporaneously a cocktail of protease
and phosphatase inhibitors diluted 100-fold (Roche, Protease and
Phosphatase Inhibitor Cocktail, reference 1836153 and
104906837001, respectively). Protein concentrations in each sample
were determined using microBCA technique according to manufac-
turer’s instructions (MicroBCA Protein assay kit, Kit Pierce reference
23235). Western blotting was performed loading 20 μg of proteins in
each gel well; pAkt-T308 was revealed using antiphospho Akt Thr308
rabbit monoclonal antibody (Cell Signaling Technology, reference
4056) followed by an antirabbit IgG HRP conjugate antibody
(Promega, reference W401B). GAPDH was revealed in control
using anti-GAPDH rabbit monoclonal antibody 14C10 (Cell Signaling
Technology, reference 2118), followed by the same antirabbit IgG
HRP conjugate antibody. After Western blotting revelation according
to the operating procedure instructions, luminescence was read using
FujiFilm (Ray Test) apparatus. This instrument measures the total
signal of luminescence obtained on Fujifilm machine (AU) for each
selected band. Then, it subtracts the background value (BG)
proportional to the size of the selected band or area. The background
is calculated from a band taken on the specific background of the
Western blot to obtain the specific signal or (AU-BG) for each band.
The software of the machine calculates the percentage of specific
activity obtained for each band as a function of a standard signal
selected as the 100% (DMSO 0.1%), which gives the percentage of
inhibition for each compound concentration. The activity of the
product was estimated by using the IC50 obtained from a dose−
response curve with six concentrations tested in single replicate and
fitted with XLFit 4 software from Excel version 4.2.2 using the four-
parameter logistic model (eq 205). IC50 values represent the mean
from at least two independent experiments; when more than a 3-fold
difference was observed between the two independent estimations,
additional experiments were performed in order to determine the
mean IC50.
The dose was considered as therapeutically active when ΔT/ΔC
was lower than 40%.37
For statistical analysis, a two-way ANOVA with repeated measures
was applied to the ranks of the tumor volume changes from baseline
(TVday − TV0) and (CVday − CV0), followed by a Winer analysis at
each day of tumor measurement. In case of significance, a Dunnett’s
test was performed versus vehicle. This statistical analysis was
performed on SAS system release 8.2 for SUN4 via Everstat 5.0
software. A probability less than 5% (p < 0.05) was considered as
significant.37
Crystallography: Determination of p110δ Crystal Structure.
Mouse p110δ (106−1044) was expressed and purified according to
published procedures.27 The protein crystallized in presence of 6−9%
(w/v) poly ethylene glycol (PEG) 8000, 12−18% ethylene glycol, 100
mM of a carboxylic acids mixture (sodium formate, sodium citrate,
sodium oxamate, ammonium acetate, and potassium/sodium tartrate)
in 100 mM MES/imidazole buffer pH 6.5 at 20 °C, and a solution of
crystal seeds diluted to 1/200000. Crystal seeds were obtained by
crystallizing the protein in presence of 14−16% PEG 8000, 28−32%
ethylene glycol, and 100 mM carboxylic acids in 100 mM MES/
Imidazole buffer pH 6.5. Co-crystals were obtained by mixing 1 μL of
protein (10 mg/mL), 1 mM compound 8 and 1 μL of precipitant with
0.2 μL of the diluted seeding solution. Crystals were flash-frozen in
In Vivo PD Assay. CB17/lCR-Prkdc severe combined immuno-
deficiency (SCID)/Crl mice, at 6−10 weeks old, were bred at Charles
River France (Domaine des Oncins, 69210 L’Arbresle, France) from
strains obtained from Charles River, USA, and were over 17 g at start
of treatment after an acclimatization time of at least 5 days. All in vivo
protocols were approved by local ethical committee. For PK/PD
studies, animals bearing tumors were randomized according to a given
median tumor burden range on a specific day post tumor implantation,
and treatment was administered as indicated with three animals per
group. At designated time point post-treatment as indicated, mice were
anesthetized and euthanized, and blood samples were collected and
kept for PK analysis. Tumors were resected, frozen in dry ice in small
fragments in a screw-capped tube containing 2.8 mm ceramic beads
(Ozyme BER1030), and stored at −80 °C until use. Protein extracts
were prepared in lysis buffer [10 mM Tris pH 7.5 (InVitrogen 15567−
4803
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
liquid nitrogen prior to data collection. Diffraction data for p110δ-
compound 8 were collected at the European Synchrotron Radiation
Facility, beamline ID29 using λ = 0.9784 Å and a Pilatus 6 M detector
(DECTRIS Ltd.). Data were processed using autoPROC,38 and the
structure was solved by molecular replacement using MOLREP39 with
the mouse p110δ structure as the search model (PDB code 2WXF)
(Table 8). The structure was refined using BUSTER/TNT40 and
model building was performed with COOT.41 Coordinates and
structure factors were deposited at the Protein Data Bank under the
code 4AJW.
(6) Jia, S.; Liu, Z.; Zhang, S.; Liu, P.; Zhang, L.; Hyun, S.; Zhang, L.
J.; Signoretti, S.; Loda, M.; Roberts, T. M.; Zhao, J. J. Essential roles of
PI(3)K−p110β in cell growth, metabolism and tumorigenesis. Nature
2008, 45, 776−779.
(7) Lannutti, B. J.; Meadows, S. A.; Herman, S. E.; Kashishian, A.;
Steiner, B.; Johnson, A. J.; Byrd, J. C.; Tyner, J. W.; Loriaux, M. M.;
Deininger, M.; Druker, B. J.; Puri, K. D.; Ulrich, R. G.; Giese, N. A.
CAL-101, a p110delta selective phosphatidyl-inositol-3-kinase inhib-
itor for the treatment of B-cell malignancies, inhibits PI3K signaling
and cellular viability. Blood 2011, 117, 561−594.
(8) Herman, S. E.; Gordon, A. L.; Wagner, A. J.; Heerema, N. A.;
Zhao, W.; Flynn, J. M.; Jones, J.; Andritsos, L.; Puri, K. D.; Lannutti, B.
J.; Giese, N. A.; Zhang, X.; Wei, L.; Byrd, J. C.; Johnson, A. J.
Phosphatidylinositol 3-kinase-δ inhibitor CAL-101 shows promising
preclinical activity in chronic lymphocytic leukemia by antagonizing
intrinsic and extrinsic cellular survival signals. Blood 2010, 116, 2078−
88.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +33(0)1 58 93 36 14. Fax: +33(0)1 58 93 80 14. E-
mail: frank.halley@sanofi.com.
Notes
(9) Jessen, K.; Kessler, L.; Kucharski, J.; Guo, X.; Staunton, J.; Janes,
M.; Elia, M.; Banerjee, U.; Lan, L.; Wang, S.; Stewart, J.; Luzader, A.;
Darjania, L.; Li, L.; Chan, K.; Martin, M.; Ren, P.; Rommel, C.; Liu, Y.
A Potent and Selective PI3kα Inhibitor, INK1117, Targets Human
Cancers Harboring Oncogenic PIk3CA Mutations. AACR Special
Conference on Targeting PI3K/mTOR Signaling in Cancer, San
Francisco, CA, February 24−27, 2011, Poster A171.
(10) Arteaga C. L. Mechanistic basis for treatment combinations with
PI3K inhibitors. 23rd EORTC-NCI-AACR Symposium on Molecular
Targets and Cancer Therapeutics, San Francisco, CA, November 12−
16, 2011, Abstract CN03−01.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Agnes
̀
Vergezac, Laurent Bassinet, Virginie
ise Herve, Alexandre
onique Lalleman,
ic Foucault, and Chagri Boumaya
Boisrobert, Laurent Dassencourt, Franco
̧
́
Rohaut, Anne Thomas, Yvette Ruffin, Ver
Marie-France Bachelot, Freder
́
́
́
for their technical assistance. We thank Dr. Carlos Garcia-
Echeverria for his guidance and thorough proofreading of the
manuscript.
(11) Markby, D.; Rice, K.; Bussenius, J.; Jaeger, C.; Nguyen, L.;
Virone-Oddos, A.; Foster, P. Characterization of Novel Series of
Selective PI3Kα and PI3Kα/mTOR-dual inhibitors. 22nd EORTC-
NCI-AACR Symposium on Molecular Targets and Cancer Ther-
apeutics, Berlin, Germany, November 16−19, 2010, Poster 149.
(12) Virone-Oddos, A.; Halley, F.; Delorme, C.; Bonnevaux, H.;
Nicolas, J.-P.; Karlsson, A.; Abecassis, P.-Y.; Vincent, L.; Lengauer, C.
Discovery and characterization of PI3Kβ isoform-selective inhibitors.
22nd EORTC-NCI-AACR Symposium on Molecular Targets and
Cancer Therapeutics, Berlin, Germany, November 16−19, 2010,
Poster 115.
(13) Carry, J.-C.; Certal, V.; Halley, F.; Karlsson, K. A.; Schio, L.;
Thompson, F. Novel [4-(morpholin-4-yl)-6-oxo-1,6-dihydropyrimi-
din-2-yl]amide derivatives, their preparation, their pharmaceutical
compositions and their use as AKT(PKB) phosphorylation inhibitors
for treating cancers. PCT Int. Appl. WO2011001114, 2011.
(14) Certal, V.; Halley, F.; Virone-Oddos, A.; Delorme, C.; Karlsson,
DEDICATION
■
This paper is dedicated to the memory of our friend and
colleague Dominique Damour.
ABBREVIATIONS USED
■
bid, bis in die (twice a day); DCM, dichloromethane; DIPEA,
diisopropylethylamine; DMF, N,N-dimethylformamide;
DMSO, dimethyl sulfoxide; EC50, effective concentration at
half maximal effect; EDCI, 1-ethyl-3-(3-dimethylaminopropyl)-
carbodiimide; HLM, human liver microsomes; IC50, inhibitory
concentration at half-maximal effect; iv, intravenous; LE, ligand
efficiency; PAkt, phosphoserine−threonine kinase Akt; MES, 4-
morpholineethanesulfonic acid; PD, pharmacodynamic; PI3K,
phosphoinositide 3-kinase; PK, pharmacokinetic; po, per os;
PTEN, phosphatase and TENsin homologue; qd, quaque die
(once a day); SAR, structure−activity relationship; SBE-
betacyclodextrin, sulfobutyl ether β-cyclodextrin; SCID, severe
combined immunodeficiency; TFA, trifluoroacetic acid; THF,
tetrahydrofuran; xantphos, 4,5-bis(diphenylphosphino)-9,9-di-
methylxanthene
́
K. A.; Thompson, F.; Filoche-Romme, B.; El-Ahmad, Y.; Carry, J.-C.;
Abecassis, P.-Y.; Bonnevaux, H.; Nicolas, J.-P.; Morales, R.; Michot,
N.; Vade, I.; Louboutin, A.; Perron, S.; Doerflinger, G.; Tric, B.;
Monget, S.; Schio, L. Preparation and optimization of new 4-
(morpholin-4-yl)- (6-oxo-1,6-dihydropyrimidin-2-yl)amide derivatives
as PI3Kβ inhibitors. Bioorg. Med. Chem. Lett. to be submitted for
publication.
(15) Certal, V.; Filoche-Romme, B.; Halley, F.; Karlsson, K. A.;
Thompson, F. Novel 6-(morpholin-4-yl)pyrimidin-4-(3H)-one deriv-
atives, their preparation, their pharmaceutical compositions and their
use as AKT(PKB) phosphorylation inhibitors for treating cancers.
PCT Int. Appl. WO2011001115, 2011.
(16) Hopkins, A. L.; Groom, C. R.; Alex, A. Ligand efficiency: a
useful metric for lead selection. Drug Discovery Today 2004, 9, 430−
431.
(17) Robertson, A. D.; Jackson, S.; Kenche, V.; Yaip, C.; Prabaharan,
H.; Thompson, P. Therapeutic preparation and formulation of
morpholino-substituted heterocycles as phosphoinositide 3-kinase
inhibitors for therapeutic use. PCT Int. Appl. WO200153266, 2001.
(18) Jackson, S.; Robertson, A. D.; Kenche, V.; Thompson, P.;
Prabaharan, H.; Anderson, K.; Abbott, B.; Goncalves, I.; Nesbitt, W.;
Schoenwaelder, S. Preparation of morpholinyl- and pyridinyl-
substituted heterobicyclic ketones as selective inhibitors of phosphoi-
REFERENCES
■
(1) Liu, P.; Cheng, H.; Roberts, T. M.; Zhao, J. J. Targeting the
phosphoinositide 3-kinase pathway in cancer. Nature Rev. Drug
Discovery 2009, 8, 627−644.
(2) Zhao, L.; Vogt, P. K. Class 1 PI3K in oncogenic cellular
transformation. Oncogene. 2008, 27, 5486−5496.
(3) Ihle, N. T.; Powis, G. The biological effects of isoform-specific
PI3-kinase inhibition. Curr. Opin. Drug Discovery Dev. 2010, 13, 41−49.
(4) Ward, S. G.; Finan., P Isoform-specific phosphoinositide 3-kinase
inhibitors as therapeutic agents. Curr. Opin. Pharmacol. 2003, 3, 426−
434.
(5) Wee, S.; Wiederschain, D.; Maira, S.-M.; Loo, A.; Miller, C.;
deBeaumont, R.; Stegmeier, F.; Yao, Y.-M.; Lengauer, C. PTEN-
deficient cancers depend on PIK3CB. Proc. Natl. Acad. Sci. U.S.A.
2008, 105, 13057−1362.
4804
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805

Journal of Medicinal Chemistry
Article
nositide 3-kinase β for use against thrombosis. PCT Int. Appl.
WO2004016607, 2004.
(37) Plowman, J.; Dykes, D. J.; Hollingshead, M.; Simpson-Herren,
L.; Alley, M. C. Human tumor xenograft models in NCI drug
development in Anticancer Drug Development Guide: Preclinical
Screening, Clinical Trials, and Approval. Humana Press Inc.: Totowa,
NJ, 1997; pp101−125.
(38) Vonrhein, C.; Flensburg, C.; Keller, P.; Sharff, A.; Smart, O.;
Paciorek, W.; Womack, T.; Bricogne, G. Data processing and analysis
with the autoPROC toolbox. Acta Crystallogr., Sect. D: Biol. Crystallogr.
2011, 67, 293−302.
(39) Collaborative Computational Project, No. 4. The CCP4 suite:
programs for protein crystallography. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 1994, 50, 760−3.
(40) Bricogne, G. Ab initio macromolecular phasing: blueprint for an
expert system based on structure factor statistics with built-in
stereochemistry. Methods Enzymol. 1997, 277, 14−18.
(41) Emsley, P.; Cowtan, K. Coot: Model-Building Tools for
Molecular Graphics. . Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004,
12 (Part 1), 2126−2132.
(19) Walker, E. H.; Pacold, M. E.; Perisic, O.; Stephens, L.; Hawkins,
P. T.; Wymann, M. P.; Williams, R. L. Structural Determinants of
Phosphoinositide 3-Kinase Inhibition by Wortmannin, LY294002,
Quercetin, Myricetin, and Staurosporine. Mol. Cell 2000, 6, 909−919.
(20) Knight, Z. S.; Gonzalez, B.; Feldman, M. E.; Zunder, E. R.;
Goldenberg, D. D.; Williams, O.; Loewith, R.; Stokoe, D.; Balla, A.;
Toth, B.; Balla, T.; Weiss, W. A.; Williams, R., L.; Shokat, K. M. A
Pharmacological Map of the PI3-K Family Defines a Role for p110α in
Insulin Signaling. Cell 2006, 125, 733−747.
(21) Guertin, D. A.; Sabatini, D. M. Defining the Role of mTOR in
Cancer. Cancer Cell 2007, 12, 9−22.
(22) Bissantz, C.; Kuhn, B.; Stahl, M. A Medicinal Chemist’s Guide
to Molecular Interactions. J. Med. Chem. 2010, 53, 5061−5084.
(23) Battle, G. M.; Allen, F. H. Learning about Intermolecular
Interactions from the Cambridge Structural Database. J. Chem. Educ.
2012, 89, 38−44.
(24) Equilibrated Solubility Protocol: 1 mg of solid was mixed with
500 μL of 50 mM phosphate buffer at pH7.4 and the suspension was
set on a bottle roller at 60 rpm for 16 h at ambient temperature. The
mixture was then filtered and the solution analyzed by HPLC for
quantization.
́
(25) Caroa, I.; Boulenca, X.; Roussetb, M.; Meuniera, V.; Bourriea,
M.; Juliana, B.; Joyeuxc, H.; Roquesa, C.; Bergera, Y.; Zweibaumb, A.;
Fabre, G. Characterisation of a newly isolated Caco-2 clone (TC-7), as
a model of transport processes and biotransformation of drugs. Int. J.
Pharm. 1995, 116, 147−158.
(26) Caliper-based assay on 192 kinases: 79 RTK, 3 STK, 11 STE, 6
CK1, 29 AGC, 37 CAMK, 20 CGMC, 7 Atypical. After a
preincubation time of 15 min between the enzyme and the inhibitor,
the enzymatic incubations were performed at 28 °C during 90 min in a
phosphate buffer at pH 7.4 containing 1 mM DTT, then stopped by
EDTA addition. Compound 8 was tested over 3 concentrations (0.1,
1, and 10 μM).
(27) Berndt, A.; Miller, S; Williams, O.; Le, D. D.; Houseman, B. T.;
Pacold, J. I.; Gorrec, F.; Hon, W-C; Ren, P.; Rommel, C.; Gaillard, P.;
Ruckle, T.; Schwartz, K. M.; Shokat, K. M.; Shaw, J. P.; Williams, R. L.
̈
The p110δ structure: mechanisms for selectivity and potency of new
PI(3)K inhibitors. Nature Chem. Biol. 2010, 6, 117−124.
(28) Knight, Z. S.; Chiang, G. G.; Alaimo, P. J.; Kenski, D. M.; Ho, C.
B.; Coan, K.; Abraham, R. T.; Shokat, K. M. Isoform-specific
phosphoinositide 3-kinase inhibitors from an arylmorpholine scaffold
Bioorg. Med. Chem. Let. 2004, 12, 4749−4759.
(29) Alaimo, P. J.; Knight, Z. A; Shokat, K. M. Targeting the
gatekeeper residue in phosphoinositide 3-kinases. Bioorg. Med. Chem.
2005, 13, 2825−2836.
(30) Zhang, X.; Vadas, O.; Perisic, O.; Anderson, K. E.; Clark, J.;
Hawkins, P. T.; Stephens, L. R.; Williams, G. L. Structure of Lipid
Kinase p110b/p85b Elucidates an Unusual SH2-Domain-Mediated
Inhibitory Mechanism. Mol. Cell 2011, 41, 567−578.
(31) Meyer, H.; Bossert, F.; Horstmann, H. Dihydropyridines, III.
Synthesis of 2-aminodihydropyridines via Michael addition. Liebigs
Ann. Chem. 1977, 11−12, 1895−1908.
(32) Cocco, M. T.; Congiu, C.; Maccioni, A.; Plumitallo, A.; Schivo,
M. L.; Palmieri, G. Synthesis and biological activity of some pyrrole
derivatives. Farmaco, Ed. Sci. 1988, 43, 103−12.
(33) Lal, B.; D’Sa, A. S.; Kulkarni, B. K.; De Souza, N. J. A convenient
one-pot entry into novel 2-substituted-6,7-dihydro-4H-pyrimido[2,1-
a]isoquinolin-4-ones. Tetrahedron 1990, 46, 1323−1330.
(34) Kantlehner, W. Product subclass 17: 1,1-Bis(nitrogen-function-
alized) alk-1-enes: alk-1-ene-1,1-diamines. Sci. Synth. 2006, 24, 571−
705.
(35) Landwehr, J.; George, S.; Karg, E.-M.; Poeckel, D.; Steinhilber,
D.; Troschuetz, R.; Werz, O. Design and Synthesis of Novel 2-Amino-
5-hydroxyindole Derivatives That Inhibit Human 5-Lipoxygenase. J.
Med. Chem. 2006, 49, 4327−4332.
(36) About the ATCC−LGC Standards Partnership; LGC Standards:
Teddington, UK; http://lgcstandards-atcc.org/.
4805
dx.doi.org/10.1021/jm300241b | J. Med. Chem. 2012, 55, 4788−4805