Discovery of novel chiral diazepines as bombesin receptor subtype-3 (BRS-3) agonists with low brain penetration

Article abstract of DOI:10.1016/j.bmcl.2013.12.106

The discovery and optimization of a novel series of BRS-3 agonists are described. We explored a potent BRS-3 agonist with low brain penetration to avoid an adverse effect derived from central nervous system exposure. Through the derivatization process, chiral diazepines 9f and 9g were identified as possessing low brain penetration as well as potent in vitro activity against human and mouse BRS-3s.

Full text of DOI:10.1016/j.bmcl.2013.12.106

Bioorganic & Medicinal Chemistry Letters 24 (2014) 750–755
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of novel chiral diazepines as bombesin receptor subtype-3
(BRS-3) agonists with low brain penetration
a
a
a
a
Tetsuyoshi Matsufuji a, , Kousei Shimada , Shozo Kobayashi , Asuka Kawamura , Teppei Fujimoto ,
Tsuyoshi Arita a, Takashi Hara b, Masahiro Konishi b, Rie Abe-Ohya b, Masanori Izumi b,
Yoshitaka Sogawa b, Youko Nagai c, Kazuhiro Yoshida c, Hisashi Takahashi a
⇑
a Medicinal Chemistry Research Laboratories, Daiichi Sankyo Co., Ltd, 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
b Cardiovascular Metabolics Research Laboratories, Daiichi Sankyo Co., Ltd, 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
c Drug Metabolism & Pharmacokinetics Research Laboratories, Daiichi Sankyo Co., Ltd, 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The discovery and optimization of a novel series of BRS-3 agonists are described. We explored a potent
BRS-3 agonist with low brain penetration to avoid an adverse effect derived from central nervous system
exposure. Through the derivatization process, chiral diazepines 9f and 9g were identified as possessing
low brain penetration as well as potent in vitro activity against human and mouse BRS-3s.
Ó 2014 Elsevier Ltd. All rights reserved.
Received 18 November 2013
Revised 17 December 2013
Accepted 25 December 2013
Available online 2 January 2014
Keywords:
Bombesin receptor subtype-3 (BRS-3)
agonists
Diazepine
Brain penetration
Bombesin receptor subtype-3 (BRS-3) is known as an orphan G-
protein coupled receptor and belongs to the bombesin receptor
family. BRS-3 receptor is expressed not only in brain, but also in
other peripheral organs such as intestine, liver, lung, testes and
pancreas.1 Since BRS-3 deficient mice develop obesity and im-
paired glycemic regulation,2 discovering ligands that modulate
BRS-3 signaling is attractive for treatment of diabetes and obesity.
Recently, Guan et al. have reported a potent and brain penetrant
BRS-3 agonist as a clinical candidate (MK-5046),3 which showed
anti-obesity effects in rats and dogs. However, MK-5046 also
caused adverse effects, such as an increase of body temperature,
heart rate and blood pressure in animals.4 Moreover, it raised at
least blood pressure in humans in the clinical trial.5 We speculated
that these adverse effects were due to the activation of sympa-
thetic nervous system by stimulating BRS-3 in the central nervous
system (CNS), and therefore low exposure of compound to the CNS
is expected to avoid them. On the other hand, since BRS-3 is also
expressed in the peripheral tissues, there is the potential for some
anti-obesity effects by stimulating peripheral BRS-3. In this paper,
we describe the discovery and optimization of a series of BRS-3
agonists with low brain penetration.
identified by a high throughput screening campaign of our corpo-
rate library, and a benzodiapzepine sulfonamide 2 which was pre-
viously reported as a potent BRS-3 agonist MK-7725 (Fig. 1).6,7
Inspired by a structural similarity between our lead compounds
1 and the sulfonamide 2, we surmised the sulfonyl group of 2 could
be replaced to another linker and the benzene ring labeled with C
in Fig. 1 could be removed from the rigid benzodiazepine structure
(Table 1).
For an efficient derivatization, we initiated the SAR study by
adopting 4-tert-butylphenyl group by reference to the previously
reported compound 3a as an (S)-atropisomer by Liu et al.8
Although replacement of sulfonyl group with arylacyl group 3b re-
sulted in no activity, arylacetyl group 3c showed a good human
and mouse in vitro activities. The replacement tert-butyl group
with phenoxy group 3d resulted in almost no in vitro activities.
However, the synthesis of meta-phenoxy compound 3e by refer-
ence to a partial structure of 1b led to dramatically improve its
activities. Encouraged by the result, we continued further modifi-
cations of compound 3e by reducing the phenyl ring to improve
its solubility. Consequently, the introduction of 2-(3-phenoxyphe-
nyl)acetyl group as the side chain of 3e turned out to enhance the
in vitro activity against both species (3f; human EC50 = 8.5 nM,
mouse EC50 = 12.2 nM). On the other hand, conversion to the urea
group 3g provided a sharp loss of potency unlike hit compound 1b.
Derivatization of the right terminal phenyl ring of 3f was next
focused on, as described in Table 2 because this part was crucial
We first designed and synthesized a novel benzodiazepine
BRS-3 agonist derived from hit compounds 1a and 1b which were
⇑
Corresponding author.
E-mail address: matsufuji.tetsuyoshi.nf@daiichisankyo.co.jp (T. Matsufuji).
0960-894X/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.12.106

T. Matsufuji et al. / Bioorg. Med. Chem. Lett. 24 (2014) 750–755
751
O
F3CO
R
N
O
N
SO2
OH
N
N
N
B
A
N
C
N
H
F3C
N
O
1a (R = Me)
1b (R =
)
2 (MK-7725)
N
H
hEC50 = 916 nM
mEC50 = 256 nM
hEC50 = 131 nM
mEC50 = 34 nM
hEC50 = 1.4 nM
mEC50 = 22 nM
Figure 1. Our HTS hit compounds 1a and 1b and brain penetrant benzodiazepine sulfonamide 2 (MK-7725) with BRS-3 EC50 values.
Table 1
EC50 values of BRS-3 agonists 3a–3f
R
N
7
8
Ar
F3C
N
N
H
9
10
a
a
Compound
R
Ar
Human BRS-3 EC50 (nM)
Mouse BRS-3 EC50 (nM)
O
S
O
3a ((S)-isomer)
7,8-Dimethylphenyl
Phenyl
97b
33b
t-Bu
t-Bu
O
O
3b
>10000
>10000
t-Bu
3c
Phenyl
Phenyl
48
227
O
O
O
O
O
3d
368
>10000
3e
3f
Phenyl
None
25
46
O
O
O
8.5
12.2
3g
None
4378
1378
N
H
a
Data are averages of at least 3 times repeated measurement.
Data are cited from Ref. 8a.
b
for in vitro activity, and could be efficiently converted by amida-
tion of the diazepine moiety with various carboxylic acids. Also
3c with 4-tert-butylphenyl group was relatively potent, but we
did not derivatize it further in the light of its rigidity, lipophilicity
and synthetic variety. An increase of potency was observed by the
installment of a fluoro atom on the right phenyl ring (4a–4c).
Among them, the para-fluoro substituent 4c showed the best effi-
cacy, and para-tolyl substituent 4d indicated further improved
activity with an increase of logD value (>4.8). Although an install-
ment of nitrogen atom on the phenyl ring reduced activities, 4f
showed the best among them. Meanwhile, the combination of
para-methyl and 3-pyridyl group 4h provided excellent activities
with a modest lipophilicity (logD = 3.6). We also examined a
replacement of the R1 part with aliphatic groups. Relative small al-
kyl derivatives 4i–4k decreased the activities presumably because
pharmacophore of the right terminal moiety required a certain size
of lipophilic functional group. As we expected, an in vitro activity
was retained in case that the number of carbon was more than four
(4l–4p). These results suggested 4h should be balanced in terms of
activity and lipophilicity at this stage.
The preparation of diazepine derivatives listed in Tables 1 and 2
was shown in Scheme 1. The synthesis of 3b–3e was performed via
a tricyclic benzodiazepine in the usual manner.8a Subsequent ami-
dation with the corresponding carboxylic acid provided the desired
compounds 3b–3e. New scaffold compounds (3f, 4a–4h) were syn-
thesized after synthetic examination for a formation of 7-mem-
bered diazepine ring. Initially, we tried to obtain the diazepine 7
by reduction of a corresponding lactam ring, but resulted in failure
probably because of distortion of the ring. As a result, we suc-
ceeded to procure 7 from 5 by reduction of carboxyl group and
additional substitution of ethylenediamine. Then, oxidation of 6
to aldehyde directly provided 7-membered imine in a moderate
yield. A dimer imine compound was also isolated as a minor prod-
uct in this step. Finally, hydrogenation by Pd/C catalyst in ethyl

752
T. Matsufuji et al. / Bioorg. Med. Chem. Lett. 24 (2014) 750–755
Table 2
EC50 values of BRS-3 agonists 3f and 4a–4p
O
R1
N
O
F3C
N
N
H
LogDb
a
a
Compound
R
Human BRS-3 EC50 (nM)
Mouse BRS-3 EC50 (nM)
3f
Ph
8.5
6.9
6.8
3.6
1.3
254
156
977
12.2
3.5
8.7
2.7
1.0
244
106
621
4.5
4.5
4.8
4.6
>4.8
3.2
3.2
3.3
4a
4b
4c
4d
4e
4f
o-F-Ph
m-F-Ph
p-F-Ph
p-tolyl
2-Py
3-Py
4-Py
N
4g
4h
9.5
2.8
3.6
4i
4j
4k
4l
4m
Me
Et
i-Pr
n-Bu
i-Bu
2041
2046
381
59
1781
1208
569
52
NDc
NDc
NDc
NDc
4.3
70
62
4n
199
121
NDc
4o
4p
c-Pentane
c-Hexane
41
8.1
100
8.7
NDc
4.4
a
b
c
Data are averages of at least 3 times repeated measurement.
The distribution coefficients (logD) were measured between 1-octanol and phosphate buffered saline (pH 7.4).
Not determined.
e
3f
4a-4h
H
N
COOH
Cl
c, d
a, b
OH
NH2
O
F3C
N
N
H
F3C
N
N
H
F3C
N
h
f, g
N
OH
4i-4p
6
5
7
F3C
N
N
H
8
Scheme 1. Reagents and conditions: (a) BH3-THF, THF, 91%; (b) ethylenediamine (neat), 125 °C, 79%; (c) MnO2, acetone/CH2Cl2, 45%, (d) H2, 10%Pd/C, AcOEt, 99%; (e)
arylacetic acid, HATU, (i-Pr)2EtN, CH2Cl2, 72–95%; (f) [3-(benzyloxy)phenyl]acetic acid, (i-Pr)2EtN, HATU, CH2Cl2, 99%; (g) H2, 10%Pd/C, EtOH, 99%; (h) alkylhalide, K2CO3, DMF,
31–83%. HATU: (Dimethylamino)-N,N-dimethyl(3H-[1,2,3]triazolo[4,5-b]pyridin-3-yloxy)methaniminium hexafluorophosphate.
Table 3
EC50 values of BRS-3 agonists 3f and 9a–9g
R2
O
R1
N
O
F3C
N
N
H
LogDb
a
a
Compound
R1
R2
Human BRS-3 EC50 (nM)
Mouse BRS-3 EC50 (nM)
3f
9a
9b
9c
Ph
Ph
Ph
Ph
Ph
Ph
Ph
p-F-Ph
i-Bu
N
H
8.5
12
1562
11
12.2
19
2260
11
15
3.4
3124
2.7
62
4.5
4.1
NDc
NDc
1.2
1.2
1.2
1.2
0.9
CH2OH
CO2H
CH2CH2OH
CH2CO2H
(R)-CH2CO2H
(S)-CH2CO2H
(R)-CH2CO2H
(R)-CH2CO2H
9d
13
(R)-9d
(S)-9d
9e
1.6
2071
1.7
17
9f
9g
(R)-CH2CO2H
1.7
3.5
0.3
a
b
c
Data are averages of at least 3 times repeated measurement.
The distribution coefficients (logD) were measured between 1-octanol and phosphate buffered saline (pH 7.4).
Not determined.

T. Matsufuji et al. / Bioorg. Med. Chem. Lett. 24 (2014) 750–755
753
MeO
N
MeO
H
N
a-c
d, e
f, g
m-r
h
N
H
9a
9b
NHBoc
NHBoc
F3C
Cl
F3C
N
Cl
F3C
N
N
H
10
11
12
O
i
O
O
Boc
CHO
Cl
O
s
N
N
j, k, c
d, e, l
N
H
F3C
N
9d
F3C
N
Cl
F3C
N
F3C
N
N
H
N
H
13
14
15
16
t, s
(R)-9d + (S)-9d
Scheme 2. Reagents and conditions: (a) 2-methoxy-N-methoxy-N-methyl-acetamide, (i-Pr)2NH, n-BuLi/hexane, THF, À78 °C, 30%; (b) N-boc-ethylenediamine, Ti(Oi-Pr)4,
THF; (c) NaBH4, EtOH, 0 °C, 2 steps, 62%; (d) 4 N HCl/AcOEt, rt; (e) (i-Pr)2EtN, N-methylpyrrolidone, 150 °C; (f) arylacetic acid, (i-Pr)2EtN, HATU, CH2Cl2, 3 steps, 35%; (g) BBr3,
CH2Cl2, À78ꢀ0 °C, 15%; (h) Jones reagent, CH2Cl2, 17%; (i) n-BuLi/hexane, (i-Pr)2NH, DMF, THF, À78 °C, 95%; (j) N-boc-ethylenediamine, Na2SO4, CH2Cl2, rt, 82%; (k)
allylmagnesium bromide/Et2O, THF, 0 °C, 73%; (l) Boc2O, Et3N, CH2Cl2, 76%; (m) 2.5%OsO4/i-PrOH, N-methylmorpholine oxide, acetone/H2O, 0 °C to rt; (n) NaIO4, THF/H2O; (o)
NaClO2, NaH2PO4, isobutene, t-BuOH/H2O; (p) iodomethane, K2CO3, acetone, 4 steps, 78%; (q) 4 N HCl/AcOEt, 92%; (r) arylacetic acid, (i-Pr)2EtN, HATU, CH2Cl2, 72–99%; (s) aq.
NaOH, THF/MeOH, 88–97%; (t) Chiral separation (CHIRALPAK, IC, EtOH/hexanes = 30/70). HATU: (Dimethylamino)-N,N-dimethyl(3H-[1,2,3]triazolo[4,5-b]pyridin-3-
yloxy)methaniminium hexafluorophosphate.
acetate and amidation of 7 with various carboxylic acids yielded 3f
and 4a–4h. Aliphatic derivatives 4i–4p was synthesized via SN2
reaction of the phenol compound 8 and various alkylhalides.
We next turned our attention to introduction of a polar group to
the diazepine structure to obtain a potent compound with lowered
logD value aiming for low brain penetration. After a synthetic
exploration and in vitro evaluation, we discovered an installment
of aliphatic groups to R2 position depicted in Table 3 retained
the potency. Therefore, we introduced various lengths of linkers
with polar groups to the R2 position (9a–9g) and discovered all
of derivatives except for 9b showed excellent activities. A chiral
separation was also examined by using chiral column chromatog-
raphy and yielded each pure enantiomer (R)-9d and (S)-9d.9 As a
result, it was turned out that only (R)-9d retained the in vitro activ-
ities (compound (R)-9d; human EC50 = 1.6 nM, mouse EC50 = 3.4 -
nM). The absolute configuration of (R)-9d was determined by the
asymmetric synthesis mentioned later. Further SAR studies were
pursued to improve potency and decrease of logD value by intro-
duction of potent functional groups on the R1 position and a car-
boxy unit on the R2 position. This combination enabled us to
obtain 9e–9g with a good activity and low logD value, and was
especially effective for 9g, which indicated good activities in both
species with the lowest logD value (compound 9g; human
EC50 = 1.7 nM, mouse EC50 = 3.5 nM, logD = 0.3).
Each compound listed in Table 3 was synthesized from a com-
mercially available pyridine 10 as shown in Scheme 2. Similar to
the synthetic route in Scheme 1, all derivatives were synthesized
by amidation of the corresponding diazepines and carboxylic acids
after cyclization of the substituted diamines. As for 9a and 9b,
addition of lithiated 10 to the Weinreb amide, imine formation
with N-boc protected diamine, and subsequent reduction yielded
11. Then, deprotection of Boc group and cyclization under heating
condition provided the diazepine 12. Finally, amidation and depro-
tection of methyl ether were conducted to generate 9a. Among
several oxidative conditions, only Jones oxidation was successful
to afford 9b.9 Next, as a first step to install an acetic acid group
to the R1 position, aldehyde 13 was reacted with allyl Grignard re-
agent, and the resulting alcohol was oxidated to the ketone and
condensed with the diamine followed by cyclization and Boc pro-
tection to afford 15. Then, the allyl group of 15 was converted to
methyl ester compound 16 after oxidative cleavage to aldehyde,
Klaus oxidation reaction to carboxylic acid,10 and methylation of
the carboxy group. Compound 9d was procured by deprotection
of Boc group and amidation. After chiral separation of 16 by utiliz-
ing of chiral column chromatography, subsequent hydrolysis of
methyl ester afforded each isomer (R)-9d and (S)-9d.11
A synthetic route of chiral diazepine derivatives 9e–9g was
shown in Scheme 3. First, Ellman’s chiral sulfinamide ((S)-t-
BuS(O)NH2) was adopted to stereospecifically introduce an allyl
group to the R1 position.12 After imine formation with the alde-
hyde 13 and the Ellman’s sulfinamide, an effective asymmetric
addition of allyl group to the imine moiety was carried out under
the condition using Indium catalyst in aqueous NaBr, which was
reported by Sun et al.13 A pure desired (R,S)-diastereomer 17 was
easily isolated by column chromatography in a high yield.14
Replacement of the sulfinamide with N-Boc ethylenediamine after
several steps, deprotection of Boc group and cyclization yielded a
key chiral intermediate 19.15 Finally, the desired compounds 9e–
9g were synthesized by use of 19 in the same manner illustrated
in Scheme 2.16,17
Pharmacokinetic (PK) parameters in mice of compounds 3f, 9f
and 9g were evaluated (Table 4). Highly lipophilic compound 3f
was not enough exposed to blood due to low solubility and micro-
somal stability (MS) and it showed an undesired high brain pene-
tration in mice. On the other hand, the solubility and unbound
fractions (PB free) of diazepine derivatives 9f and 9g with a chiral
carboxyl group were dramatically improved. As a result, blood
exposure of 9f and 9g in mice was also improved. Moreover, the
O
O
HO
NHBoc
e, f
O
O
S
Boc
O
Boc
a, b
c, d
g-j
e, k, l
R
N
N
13
N
O
N
H
N
H
F3C
N
F3C
N
F3C
N
Cl
17
F3C
N
Cl
N
H
19
N
H
20
F3C
N
N
H
9e-9g
18
Scheme 3. Reagents and conditions: (a) (S)-t-BuS(O)NH2, Ti(OEt)4, THF, 70 °C, 83%; (b) allylbromide, In, aq NaBr, 93%; (c) HCl/dioxane,MeOH, 82%; (d) tert-butyl N-(2-
oxoethyl)carbamate, NaBH(OAc)3, dichloroethane, quant.; (e) 4 N HCl/AcOEt, rt; (f) (i-Pr)2EtN, N-methylpyrrolidone, 160 °C, then (Boc)2O, 2 steps, 72%; (g) 2.5%OsO4/i-PrOH,
N-methylmorpholine oxide, acetone/H2O, 0 °C to rt; (h) NaIO4, THF/H2O; (i) NaClO2, NaH2PO4, isobutene, t-BuOH/H2O; (j) TMSCH2N2, THF/MeOH, rt, 4 steps, 73%; (k) 1-
arylacetic acid, (i-Pr)2EtN, HATU, CH2Cl2, 74–95%; (l) aq NaOH, THF/MeOH, 90–96%.

754
T. Matsufuji et al. / Bioorg. Med. Chem. Lett. 24 (2014) 750–755
Table 4
Physical properties and mouse PK parameters of 3f, 9f and 9g
c
c
Compound
In vitro EC50 (nM)
LogDb
Solubility JP1/JP2
g/mL)
PB freea (%)
Cmax
g/mL)
AUCc (hr⁄
l
g mL)
T1/2 (h)
MSd (%)
PSAe
Mouse Kp brainf
(
l
(l
3f
9f
9g
12
62
3.5
4.5
0.9
0.3
11/5.1
40/810
960/950
0.67
3.3
4.6
0.060
0.36
0.090
0.041
0.39
0.20
0.82
3.7
2.6
0.70
70
94
53
88
98
0.72g
0.043g
0.002g,h
a
b
c
d
e
f
Unbound fractions (%) in mouse plasma. PB: Protein Binding.
The distribution coefficients (logD) were measured between 1-octanol and phosphate buffered saline (pH 7.4).
Average of 2 mice dosed at 10 mg/kg po. Each dose was administered by DMA/PG/(20%HPbCD/saline): 10/10/80 as a solvent.
Remaining (%) of the tested compound after 0.5 h incubation with mouse liver microsome.
Polar surfice area (Å2).
Kp brain: brain to plasma concentration ratio.
Kp value was determined after a single administration of compound at 30 mg/kg po to mice. The measurement of each concentration was conducted at 2.5 h after
g
administration.
Measured by racemic compound of 9g.
h
7. We confirmed a binding ability of some key compounds to human BRS-3
brain to plasma concentration ratios (Kp brain) of them, especially
9g, were fairly low in parallel with their logD values as we ex-
pected. In addition, their high polar surface area (PSA) value also
might contribute their low brain exposure because high PSA is sup-
posed to lower blood–brain barrier penetration.18 Thus, we found
out two promising candidates 9f and 9g as BRS-3 agonists with
high in vitro activity, blood exposure and low Kp brain value.
In summary, we obtained chiral diazepine derivatives as novel
BRS-3 agonists with low brain penetration. Removal of the benzene
ring labeled with C of 2, which was suggested from our hit com-
pounds 1, resulted in a first breakthrough to obtain an original
scaffold. Introduction of a polar group to the diazepine structure
provided the potent derivative 9d. By the chiral separation of 9d,
we discovered only (R)-9d was highly active. Also an effective
asymmetric synthesis of the desired (R)-enantiomers 9e–9g was
achieved. Finally, various polar compounds were obtained by com-
bining each preferable functional group on the R1 and R2 position.
This strategy enabled us to obtain chiral diazepines 9f and 9g with
potent in vitro BRS-3 activity, good blood exposure and low brain
penetration. These compounds are expected to avoid adverse ef-
fects derived from CNS, hence in vivo efficacy and safety evalua-
tions of them are underway and to be reported in near future.
receptor by a radioligand receptor binding assay (data not shown).
8. (a) Xiao, -M. G.; Joseph, M. M.; Liming, Y.; Kate, A. R.; Sheng, -P. W.; Stephanie,
K. S.; Jennifer, A. K.; Hong, Y.; Lauren, P. S.; Terry, D. F.; Oksana, P.; Yanqing, K.;
Theresa, M. K.; Iyassu, S.; Linus, S. L.; Jasminka, D.; Kathy, A. L.; Stephanie, C.;
Ravi, P. N.; Donald, J. M.; Alison, M. S.; Marc, L. R. ACS Med. Chem. Lett. 2011, 2,
933; (b) Harry, R. C.; Yan, G.; Ping, L.; Marc, C.; Thomas, J. L., Jr.; Linda, C.;
Theresa, M. K.; Yanqing, K.; Oksana, P.; Xiao-M, G.; Donald, J. M.; Joseph, M. M.;
Judith, N. G.; Kate, R.; Sheng, -P. W.; Alison, M. S.; Randy, M.; Jianmei, P.; Maria,
M.; Kathy, L.; Jasminka, D.; Marc, L. R.; Ravi, P. N.; Linus, S. L. ACS Med. Chem.
Lett. 2012, 3, 252; (c) Harry, R. C.; Yan, G.; Ping, L.; Thomas, J. L., Jr.; Marc, C.;
Linda, C.; Theresa, M. K.; Yanqing, K.; Oksana, P.; Xiao-M, G.; Donald, J. M.;
Joseph, M. M.; Kate, R.; Sheng, -P. W.; Alison, M. S.; Judith, N. G.; Randy, M.;
Jianmei, P.; Kathy, L.; Jasminka, D.; Jian, G. N.; Wes, A. S.; Christopher, J. W.;
Xiaoyi, G.; Ying, -D. G.; Viktor, H.; Marc, L. R.; Ravi, P. N.; Linus, S. L. Bioorg. Med.
Chem. 2012, 20, 2845.
9. Bowden, K.; Heilbron, I. M.; Jones, E. R. H. J. Chem. Soc. 1946, 39.
10. Kraus, G. A.; Roth, B. J. Org. Chem. 1980, 45, 4825.
11. Each enantiomer (R)-9d and (S)-9d was obtained by hydrolysis after chiral
HPLC separation of 15: column, Chiral Pack IC (4.6 Â 250 mm); eluent, ethanol/
hexanes = 30/70. The ee value of the methyl esters of (R)-9d and (S)-9d were
determined by chiral HPLC analysis: column, Chiral Pack IC (4.6 Â 250 mm);
flow rate 1.0 mL/min; temperature, 27 °C; tR of (R)-enantiomer = 6.56 min; tR
of (S)-enantiomer = 7.47 min. The retention time of the methyl ester of (R)-
enantiomer ((R)-9d) via chiral separation was well corresponded to the one
sythesized by the stereospecific route in Scheme 3.
12. (a) Liu, G.; Cogan, D. A.; Ellman, J. A. J. Am. Chem. Soc. 1997, 119, 9913; (b)
Ellman, J. A.; Owens, T. D.; Tang, T. P. Acc. Chem. Res. 2002, 35, 984.
13. Sun, X.-W.; Liu, M.; Xu, M.-H.; Lin, G.-Q. Org. Lett. 2008, 10, 1259.
14. An undesired (S,S)-diastereomer was also obtained in 2.0% yield.
15. The ee value of the key intermediate 19 was determined to 98.1% ee by chiral
HPLC analysis: column, Chiral Pack IA (4.6 Â 150 mm); eluent, ethanol/
isopropylalcohols = 50/50; flow rate, 1.0 mL/min; temperature, 27 °C; tR of
(R)-enantiomer = 3.56 min; tR of (S)-enantiomer = 4.11 min. The absolute
configration of 19 was inferred by the reaction mechanism proposed by the
Ellman’s papers (see Ref. 12).
Reference and notes
1. Liu, J.; Lao, Z. J.; Zhang, J.; Schaeffer, M. T.; Jiang, M. M.; Guan, X-M.; Van der
Ploeg, L. H. T.; Fong, T. M. Biochemistry 2002, 41, 8954.
2. Ohki, H.; Watase, K.; Yamamoto, K.; Ogura, H.; Yamano, M.; Yamada, K.; Maeno,
H.; Imaki, J.; Kikuyama, S.; Wada, E. Nature 1997, 390, 165.
16. The synthesis of 9f was performed as follows: Step 1. Synthesis of
N-{(1R)-1-[2-chloro-6-(trifluoromethyl)pyridin-3-yl]but-3-en-1-yl}-2-
methylpropane-2-sulfinamide (17): A mixture of 2-chloro-6-(trifluoromethyl)
pyridine-3-carbaldehyde 13 (19.0 g, 90.8 mmol), Ti(OEt)4 (41.4 g. 182 mmol),
and (S)-(À)-tert-butylsulfinamide (11.2 g. 92.6 mmol) in THF (200 mL) was
stirred at reflux for 5 h. The cooled reaction mixture was concentrated, and
dissolved in ethylacetate (500 mL). The resulting suspension was filtered
through a Celite pad, and the filtrate was concentrated and purified by column
chromatography (hexane/EtOAc = 4:1) to provide the imine (28.9 g) as a light-
yellow solid. A mixture of the imine (28.9 g, 92.5 mmol), allyl bromide (55.9 g,
0.46 mmol), and indium (42.5 g, 0.37 mmol) was vigorously stirred in sat. NaBr
aq at room temperature overnight. After addition of sat. NaHCO3 aq (500 mL),
the reaction mixture was extracted with ethylacetate (250 mL Â 3). The
3. Iyassu, K. S.; Christopher, F.; Michael, M.-C. L.; David, C.; James, P. J.; Randy, M.;
Jianmei, P.; Oksana, P.; Yanqing, K.; Theresa, M. K.; Xiao, -M. G.; Donald, J. M.;
Jennifer, A. K.; Joseph, M. M.; Kathryn, L.; Jasminka, D.; Peter, R. G.; Alan, J. H.;
Marc, L. R.; Ravi, P. N.; Matthew, J. W.; Lin, L. S. ACS Med. Chem. Lett. 2011, 2, 43.
4. Guan, X.-M.; Metzger, J. M.; Yang, L.; Raustad, K. A.; Wang, S.-P.; Spann, S. K.;
Kosinski, J. A.; Yu, H.; Shearman, L. P.; Faidley, T. D.; Palyha, O.; Kan, Y.; Kelly, T.
M.; Sebhat, I.; Lin, L. S.; Dragovic, J.; Lyons, K. A.; Craw, S.; Nargund, R. P.;
Marsh, D. J.; Strack, A. M.; Reitman, M. L. J. Pharmacol. Exp. Ther. 2011, 336, 356.
5. Reitman, M. L.; Dishy, V.; Moreau, A.; Denney, W. S.; Liu, C.; Kraft, W. K.; Mejia,
A. V.; Matson, M.; Stoch, S. A.; Wagner, J. A.; Lai, E. J. Clin. Pharmacol. 2012, 52,
1306.
6. The 384-well IP-One HTRFÒ Assay (Cisbio, Bedford, MA) was performed as
described by the manufacturer’s protocol. CHO-K1 stably expressing human
organic layers were filtered through
a Celite pad, and dried (Na2SO4),
and mouse BRS-3 was plated at 200,000 cells per well in 50
incubated in the CO2 incubator at 37 °C for 24 h. On the next day, the media
were removed and 25 l of compounds diluted in HAM containing 0.1% BSA
ll HAM and
concentrated and purified by column chromatography (hexane/EtOAc = 1:4
with 5% of Et3N) to provide 17 as a white solid (26.0 g, 77%). 1H NMR (500 MHz,
CDCl3) d 7.99 (1H, d, J = 7.8 Hz), 7.67 (1H, d, J = 7.8 Hz), 5.81–5.73 (1H, m),
5.27 (2H, dd, J = 13.2, 17.6 Hz), 5.06–5.02 (1H, m), 3.84 (1H, d, J = 2.4 Hz), 2.82–
2.77 (1H, m), 2.53–2.47 (1H, m), 1.27 (9H, s). Step 2. Synthesis of tert-butyl[2-
({(1R)-1-[2-chloro-6-(trifluoromethyl)pyridin-3-yl]but-3-en-1-yl}amino)ethyl
carbamate (18): 17 (22.8 g, 64.3 mmol) was dissolved in 4 N HCl/MeOH
(100 mL) and stirred at room temperature for 2 h. After the reaction mixture
was concentrated, the residue was added to diethylether (300 mL). The
resulting suspension was filtered, and dried in vacuo to provide the amine
HCl salt (15.1 g, 82%) as a white solid. A mixture of the amine HCl salt (15.1 g,
52.8 mmol), tert-butyl N-(2-oxoethyl)carbamate (12.6 g, 79.3 mmol), and
NaBH(OAc)3 (25.2 g, 95.1 mmol) in CH2Cl2 (250 mL) was stirred at room
temperature for 1 h, then warmed to room temperature for 4 h. After addition
l
and 20 mM LiCl were added to each well and serial diluted IP1 standards
(Cisbio) were also added to corresponding wells at this step. After the cells
were incubated for 1 h in the CO2 incubator at 37 °C, all the media were
removed again and d2-labeled IP1 and cryptate-labeled anti-IP1 monoclonal
antibody diluted in lysis buffer were added sequentially. The assay plates were
kept in dark at 4 °C overnight. Ratiometic measurements of fluorescence
emission at 665 nm and 620 nm were obtained using a RUBYstar fluorometer
(BMG Labtech, Ortenberg, Germany). IP1 levels in each well were calculated
according to the standard curves on each plate. EC50 values were obtained by
fitting data to a nonlinear curve-fitting program (GraphPad Software, Inc., La
Jolla CA).

T. Matsufuji et al. / Bioorg. Med. Chem. Lett. 24 (2014) 750–755
755
of sat.NaHCO3 aq (300 mL), the reaction mixture was extracted with CH2Cl2
(250 mL Â 2). The organic layers were washed with brine, dried (Na2SO4),
concentrated and purified by column chromatography (hexane/EtOAc = 2:5) to
provide 18 as a white solid (21.6 g, quant.). 1H NMR (500 MHz, CDCl3) d 8.09
(1H, d, J = 7.8 Hz), 7.60 (1H, d, J = 8.2 Hz), 5.78–5.68 (1H, m), 5.14–5.07 (2H, m),
4.69 (1H, s), 4.15 (1H, dd, J = 7.8, 3.9 Hz), 4.09 (1H, dd, J = 6.6, 14.5 Hz), 3.15
(2H, br s), 2.61–2.56 (1H, m), 2.53–2.49 (1H, m), 2.44–2.38 (1H, m), 2.21–2.13
(1H, m), 1.18 (9H, d, J = 2.0 Hz). Step 3. Synthesis of tert-butyl (5R)-5-(prop-2-
en-1-yl)-8-(trifluoromethyl)-1,2,3,5-tetrahydro-4H-pyrido[2,3-
concentrated and purified by column chromatography (EtOAc/hexane = 1:3)
to provide 20 as colorless oil (8.40 g, 73%). 1H NMR (500 MHz, CDCl3): a
mixture of conformers d 7.76–7.61 (1H, m), 7.16 (1H, d, J = 7.8 Hz), 5.70 (1H, d,
J = 90.8 Hz), 5.03 (1H, s), 4.14–4.04 (1H, m), 3.67 (3H, s), 3.51–3.44 (1H, m),
3.37 (1H, d, J = 14.6 Hz), 3.21–3.06 (2H, m), 2.95 (1H, dd, J = 15.6, 8.3 Hz), 1.36
(9H, s). Step 5. Synthesis of [(5R)-4-({3-[(6-methylpyridin-3-yl)oxy]phenyl}
acetyl)-8-(trifluoromethyl)-2,3,4,5-tetrahydro-1H-pyrido[2,3-e][1,4]diazepin-
5-yl]acetic acid (9f): 20 (8.82 g, 22.7 mmol) was dissolved in 4 N HCl/1,4-
dioxane (50 mL) and stirred at room temperature for 1 h. (i-Pr)2O (300 mL) was
added to the mixture. The deposited hydrochloride salt was separated by
filtration and dried in vacuo to provide the diazepine as white solid (7.1 g). A
mixture of the diazepine (2.00 g, 4.38 mmol), 2-(3-isobutoxyphenyl)acetic acid
(1.05 g, 5.04 mmol), (i-Pr)2EtN (3.05 mL, 17.5 mmol), HATU (2.08 g, 5.48 mmol)
in CH2Cl2 (20 mL) was stirred at room temperature overnight. The mixture was
diluted in CH2Cl2 (100 mL), washed with sat. NaHCO3 aq and brine, dried
(Na2SO4), concentrated and purified by column chromatography (EtOAc/
hexane = 1:1) to provide the acylated diazepine as colorless oil (2.36 g). A
mixture of the acylated diazepine (2.26 g, 4.07 mmol), 2 N NaOH aq (40 mL) in
THF/MeOH(20 mL/20 mL) was stirred at room temperature for 2 h. The
mixture was acidified with 2 N HCl aq and extracted with EtOAc. The organic
layer was washed with sat. NaHCO3 aq and brine, dried (Na2SO4) and
concentrated in vacuo. The obtained residue was triturated in Et2O/hexane,
filtered and dried in vacuo to provide 9f as white solid (1.69 g, 89%) 1H NMR
(400 MHz, CDCl3): a mixture of conformers d 7.81 (0.5H, d, J = 7.4 Hz), 7.21–
7.13 (1.5H, m), 6.90 (1H, dd, J = 26.6, 7.4 Hz), 6.79–6.71 (3H, m), 6.21 (0.5H, t,
J = 7.0 Hz), 5.46 (0.5H, t, J = 7.8 Hz), 5.16–5.01 (1H, m), 3.99 (1H, dd, J = 28.8,
15.5 Hz), 3.87-3.54 (5H, m), 3.47–3.22 (2H, m), 3.19–2.95 (2H, m), 2.88–2.75
(1H, m), 2.07–1.97 (1H, m), 1.01–0.97 (6H, m). MS (ESI) m/z: 466 (M+H)+.
HRMS (ESI) m/z: 464.1793 (calcd for C23H26N3O4F3: 464.1797). a2D1:0 À40.2 (c
1.00, CHCl3).
e][1,4]diazepine-4-carboxylate (19): 18 (21.6 g, 54.8 mmol) was dissolved in
4 N HCl/ethylacetate (100 mL) and stirred at room temperature for 1 h. After
the reaction mixture was concentrated, the residue was dissolved in CH2Cl2
(300 mL), washed with sat. NaHCO3 aq and brine, dried (Na2SO4), concentrated
and purified by column chromatography (MeOH/CH2Cl2 = 1:12) to provide the
diamine as white solid (12.7 g). A mixture of the diamine (12.7 g. 43.3 mmol)
and (i-Pr)2EtN (23.7 mL, 130 mmol) in NMP (250 mL) was stirred at 160 °C for
8 h. Subsequently, (Boc)2O (14.2 g, 65.0 mmol) was added to the cooled
mixture. After stirring at room temperature for 1 h, The mixture was diluted
with Et2O, washed with brine, dried (Na2SO4), concentrated and purified by
column chromatography (EtOAc/hexane = 1:4) to provide 19 as brown oil
(11.1 g, 72%). 1H NMR (500 MHz, CDCl3): a mixture of conformers d 7.53–7.39
(1H, m), 7.06 (1H, d, J = 7.8 Hz), 5.64–5.54 (1H, m), 5.31–4.96 (4H, m),
4.10–3.97 (1H, m), 3.35–3.27 (2H, m), 3.10 (1H, dt, J = 24.9, 10.5 Hz), 2.80–2.68
(1H, m), 2.45 (1H, td, J = 14.9, 7.2 Hz), 1.40 (9H, s). Step 4. Synthesis of tert-
butyl (5R)-5-(2-methoxy-2-oxoethyl)-8-(trifluoromethyl)-1,2,3,5-tetrahydro-
4H-pyrido[2,3-e][1,4]diazepine-4-carboxylate (20): A mixture of 19 (11.1 g,
31.1 mmol), 2.5%OsO4 in i-PrOH (8.00 mL. 0.78 mmol), and N-methylmor-
pholine oxide (4.74 g. 40.5 mmol) in acetone/H2O (90 mL/30 mL) was stirred at
0 °C overnight. The mixture was diluted with AcOEt (300 mL), washed with
brine, dried (Na2SO4), concentrated and purified by column chromatography
(MeOH/CH2Cl2 = 1:8) to provide the diol as brown oil (12.9 g). A mixture of the
diol (12.9 g, 31.2 mmol), sodium periodate (13.3 g. 62.3 mmol) in THF/H2O
(150 mL/50 mL) was stirred at room temperature for 1.5 h. The mixture was
diluted with AcOEt (300 mL), washed with brine, dried (Na2SO4), concentrated
in vacuo to provide the aldehyde as colorless oil (11.9 g). A mixture of the
aldehyde (11.9 g, 31.1 mmol), NaH2PO4 2H2O (36.4 g. 234 mmol), NaClO
(7.04 g, 78 mmol), 2-methyl-2-butene (33.1 mL, 311 mmol) in tBuOH/H2O
(150 mL/150 mL) was stirred at room temperature overnight. After
concentration, the residue was diluted with AcOEt (300 mL Â 2), washed
with brine, dried (Na2SO4), concentrated in vacuo to provide the carboxylic
acid as brown oil (16.1 g). A mixture of the carboxylic acid (15.2 g, 22.7 mmol),
2.0 M (Trimethylsilyl)diazomethane/hexane (18.0 mL, 36.0 mmol) in THF/
MeOH (50 mL/50 mL) was stirred at 0 °C for 1 h. The mixture was
17. The synthesis of 9g was prepared in a similar manner described for 9f (See Ref.
16, Step 5). [(5R)-4-{[3-(2-methylpropoxy)phenyl]acetyl}-8-(trifluoromethyl)-
2,3,4,5-tetrahydro-1H-pyrido[2,3-e][1,4]diazepin-5-yl]acetic acid (9g):
A
white solid, 1H NMR (400 MHz, CD3OD): a mixture of conformers d 8.14 (1H,
dd, J = 16.6, 2.4 Hz), 7.78 (0.5H, d, J = 7.3 Hz), 7.35 (1H, t, J = 5.6 Hz), 7.30–7.27
(2H, m), 7.16 (0.5H, d, J = 7.3 Hz), 7.09–6.93 (3H, m), 6.85–6.72 (1H, m), 6.15
(0.5H, t, J = 7.6 Hz), 5.56 (0.5H, t, J = 7.3 Hz), 4.19–3.38 (5H, m), 3.26–2.81 (4H,
m), 2.53 (3H, d, J = 4.4 Hz). MS (ESI) m/z: 501 (M+H)+. HRMS (ESI) m/z:
499.1597 (calcd for C25H23N4O4F3: 499.1593). a2D1:0+32.9 (c 1.00, CHCl3).
18. (a) Doan, K. M. M.; Humphereys, J. E.; Webster, L. O.; Wring, S. A.; Shampine, L.
J.; Serabjit-singh, C. J.; Adkinson, K. K.; Polli, J. W. J. Pharmacol. Exp. Ther. 2002,
303, 1029; (b) Clark, D. E. Drug Discov. Today. 2003, 8, 927; (c) Pajouhesh, H.;
Lenz, G. R. NeuroRX 2005, 2, 541.