Design, synthesis, and pharmacological evaluation of 4-azolyl-benzamide derivatives as novel GPR52 agonists

Article abstract of DOI:10.1016/j.bmc.2017.03.064

G protein-coupled receptor 52 (GPR52) agonists are expected to improve the symptoms of psychiatric disorders. During exploration for a novel class of GPR52 agonists with good pharmacokinetic profiles, we synthesized 4-(3-(3-fluoro-5-(trifluoromethyl)benzyl)-5-methyl-1H-1,2,4-triazol-1-yl)-2-methylbenzamide (4u; half maximal effective concentration (EC₅₀)?=?75?nM, maximal response (E_max)?=?122%) starting from a high-throughput screening hit 3 (EC₅₀?=?470?nM, E_max?=?56%). The structural features of a reported GPR52 agonist were applied to 3, led to design 4-azolylbenzamides as novel GPR52 agonists. A structure–activity relationship study of 4-azolylbenzamide resulted in the design of the 1,2,4-triazole derivative 4u, which demonstrated excellent bioavailability in rats (F?=?53.8%). Oral administration of 4u (10?mg/kg) significantly suppressed methamphetamine-induced hyperlocomotion in mice. Thus, 4u is a promising lead compound for drug discovery research of GPR52 agonists.

Full text of DOI:10.1016/j.bmc.2017.03.064

Accepted Manuscript
Design, synthesis, and pharmacological evaluation of 4-azolyl-benzamide de-
rivatives as novel GPR52 agonists
Kazuyuki Tokumaru, Yoshiteru Ito, Izumi Nomura, Takashi Nakahata, Yuji
Shimizu, Emi Kurimoto, Kazunobu Aoyama, Kazuyoshi Aso
PII:
S0968-0896(17)30110-4
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.03.064
BMC 13662
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
18 January 2017
28 March 2017
29 March 2017
Please cite this article as: Tokumaru, K., Ito, Y., Nomura, I., Nakahata, T., Shimizu, Y., Kurimoto, E., Aoyama, K.,
Aso, K., Design, synthesis, and pharmacological evaluation of 4-azolyl-benzamide derivatives as novel GPR52
agonists, Bioorganic & Medicinal Chemistry (2017), doi: http://dx.doi.org/10.1016/j.bmc.2017.03.064
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Design, synthesis, and pharmacological evaluation of 4-azolyl-benzamide derivatives
as novel GPR52 agonists
*
Kazuyuki Tokumaru, Yoshiteru Ito , Izumi Nomura, Takashi Nakahata, Yuji Shimizu, Emi
*
Kurimoto, Kazunobu Aoyama, Kazuyoshi Aso
Pharmaceutical Research Division, Takeda Pharmaceutical Company Limited: 26-1,
Muraoka-Higashi 2-chome, Fujisawa, Kanagawa 251-8555, Japan
Corresponding
Author:
*K.A.,
Y.I.,
phone:
phone:
+81-466-32-1121,
+81-466-32-1128,
e-mail:
e-mail:
kazuyoshi.aso@takeda.com;
yoshiteru.ito@takeda.com

ABSTRACT
G protein-coupled receptor 52 (GPR52) agonists are expected to improve the symptoms of
psychiatric disorders. During exploration for a novel class of GPR52 agonists with good
pharmacokinetic profiles, we synthesized 4-(3-(3-fluoro-5-(trifluoromethyl)benzyl)-5-
methyl-1H-1,2,4-triazol-1-yl)-2-methylbenzamide
(4u;
half
maximal
effective
concentration (EC ) = 75 nM, maximal response (E ) = 122%) starting from a high-
50
max
throughput screening hit 3 (EC = 470 nM, E = 56%). The structural features of a
5
0
max
reported GPR52 agonist were applied to 3, led to design 4-azolylbenzamides as novel
GPR52 agonists. A structure–activity relationship study of 4-azolylbenzamide resulted in
the design of the 1,2,4-triazole derivative 4u, which demonstrated excellent bioavailability
in rats (F = 53.8%). Oral administration of 4u (10 mg/kg) significantly suppressed
methamphetamine-induced hyperlocomotion in mice. Thus, 4u is a promising lead
compound for drug discovery research of GPR52 agonists.
KEYWORDS
GPCR; GRP52 agonist; Methamphetamine-induced hyperlocomotion; Metabolic stability;
1
,2,4-Triazole.
ABBREVIATIONS
SAR, structure–activity relationship; cAMP, 3',5'-cyclic adenosine monophosphate; EC ,
50
half maximal effective concentration; MAP, methamphetamine; NMDA, N-methyl-D-

aspartate; PK, pharmacokinetic; CHO, Chinese hamster ovary; SEM, standard error of the
mean; AUC, area under the curve.
1
. INTRODUCTION
G protein-coupled receptors constitute one of the most common therapeutic targets in the
1, 2
central nervous system for treatment of psychiatric disorders . G protein-coupled receptor
5
2 (GPR52) gene was identified from a high-throughput genome database, and classified as
3
an orphan Gs-coupled G protein-coupled receptor . A detailed histological study of mice
tissue has revealed that GPR52 is colocalized with dopamine D2 receptor in the mesolimbic
4
system, and partially with dopamine D1 receptor in the prefrontal cortex . Moreover,
GPR52 knockout mice displayed psychosis-related behaviors, whereas GPR52 transgenic
4
mice showed anti-psychiatric behaviors . These results suggest that GPR52 agonists are
likely to alleviate the symptoms of psychiatric disorders via induction of cAMP
accumulation by canceling the dopamine D2 signal and activating the dopamine D1/N-
methyl-D-aspartate (NMDA) signal.
In our previous study, the benzothiophene derivative 1 was developed as a potent and
5
orally active GPR52 agonist , as shown in Figure 1. Oral administration of compound 1 (3
mg/kg) significantly suppressed methamphetamine-induced hyperactivity in mice without
inducing cataleptogenic effects or extrapyramidal side effects, which may result from
6
dopamine D2 blocking . During exploration for new scaffolds of GPR52 agonists, a
scaffold hopping study of compound 1 provided the thiazole derivative 2 (Figure 1), which
7
possessed excellent physicochemical profile and brain penetration . Furthermore, a unique

hit compound 3 was identified by an additional high-throughput screening campaign of our
in-house compound library.
Figure 1. Reported GPR52 agonists 1–2 and a HTS hit compound 3.
2
. DESIGN AND SYNTHESIS
.1. DESIGN
2
7
In a previous study, the thiazole derivative 2 was developed by a scaffold hopping
5
design strategy utilizing benzothiophene 1 . The 2-benzyl-4-methylthiazole scaffold in
compound 2 was identified as a highly promising structure with GPR52 agonistic activity.
Further morphing of this scaffold was considered to provide an attractive series of GPR52
agonists with potent in vitro and in vivo activities and good pharmacokinetic (PK) profiles.
To understand the functionality of each moiety, various conformations of compounds 2
8
and 3 were generated by using LowMode MD method in MOE 2015.10 . Their
superpositions are shown in Figure 2. It is noteworthy that (1) the pyrazole moiety in
compound 3 overlapped well with the thiazole moiety in compound 2; and (2) the
benzamide of 3 could replace the pyrazole carboxamide of 2. These findings led us to
combine the azolylbenzamide motif and the 2-benzyl-4-methylthiazole motif, which

resulted in the design of a novel series of azolylbenzamide derivatives (Figure 3). This
study describes, in detail, the biological activities of newly designed azolylbenzamides 4a–
u.
Figure 2. Superposed conformations obtained for compound 2–3 by LowMode MD
conformational search with the MMFF94x force field. 2 (green), 3 (orange). The picture
was generated using MOE 2015.10.
Figure 3. Design of azolylbenzamide 4a–u from the hit compound 3.
2
.2. CHEMISTRY

The synthesis of thiazole analogs (4a–r), pyrazole analog 4s, imidazole analog 4t, and
,2,4-triazole analog 4u is described in Schemes 1–5, 6, 7 and 8, respectively. Regarding
1
thiazole compounds (4a, 4c, and 4f–h), their synthesis was achieved by the conversion of
carboxylic acids 11a–e as shown in Scheme 1. Furthermore, hydrolysis of nitrile moiety
was an important reaction for the preparation of the targets (Scheme 2 for compounds 4b
and 4q–r; Scheme 3 for 4d–e). In Scheme 1, commercially available arylmethyl cyanides
5
a–c were allowed to react with diethyl dithiophosphate to afford the thioamides 6a–c.
Synthesis of the 5-bromothiazole compounds 8a–c was achieved by formation of thiazole
ring via reaction of the thioamides 6a–c with bromoacetone followed by bromination with
N-bromosuccinimide (NBS). The targeted analogs (4a–c, 4f–h, 4q, and 4r) were
synthesized from the 5-bromothiazole compounds 8a–c through the following multistep
reactions: (A) Suzuki-Miyaura coupling for introduction of benzoic ester, hydrolysis of
ester moiety, and condensation with ammonia, as shown in Scheme 1, and (B) Suzuki-
Miyaura coupling for introduction of benzonitrile and hydrolysis of nitrile moiety, as
shown in Schemes 2 and 3. The preparation of the 5-arylthiazole analogs, which possess an
ethyl group (4d) or an isopropyl group (4e) at the 2-position of the benzene ring is
described in Scheme 3. The intermediate 8a was reacted with arylboronic acid pinacol ester
to afford the phenolic analog 13. Compound 13 was converted to the
trifluoromethanesulfonate 14 followed by Suzuki-Miyaura coupling reaction with vinyl or
1
-propenylboronic acid pinacol ester, heterogeneous catalytic hydrogenation, and
hydrolysis of nitrile to afford the ethyl or isopropyl analogs, 4d and 4e.
a
Scheme 1. Preparation of 4-arylthiazole analogs 4a, 4c, 4f–h

a
Reagents and conditions: (a) diethyl dithiophosphate, HCl, EtOAc, 0°C to rt; (b)
bromoacetone, EtOH, reflux; (c) NBS, MeCN, 0°C to rt; (d) 8a, Pd(Ph P) , Na CO , DME–
3
4
2
3
water, 65–80°C; or, 8a, Pd (dba) , X-Phos, Cs CO , DME-water, 50°C; or, 8a, PdCl (Ph P) ,
2
3
2
3
2
3
2
K CO , DME-water, 85°C; (e) NaOH aq., EtOH–THF or MeOH–THF, 60–80°C; (f) EDCI,
2
3
HOBt-NH , DMF, rt.
3
a
Scheme 2. Preparation of 4-arylthiazole analogs 4b, 4q–r

a
Reagents and conditions: (a) 2-methyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-
yl)benzonitrile, Pd (dba) , X-Phos, Cs CO , DME-water, 50°C; (b) H O aq., K CO , DMSO,
2
3
2
3
2
2
2
3
0
°C to rt; or H SO , 80°C.
2 4
a
Scheme 3. Preparation of 4-arylthiazole analogs 4d–e
a
Reagents and conditions: (a) 2-hydroxy-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-
yl)benzonitrile, PdCl (Ph P) , K CO , DME–water, 85°C; (b) N-
phenylbis(trifluoromethanesulfonimide), Cs CO , DMF, rt; (c) vinylboronic acid pinacol
2
3
2
2
3
2
3
ester for 15a or isopropenylboronic acid pinacol ester for 15b, PdCl (dppf) or PdCl (Ph P) ,
2
2
3
2

K CO , DME–water, 85°C; (d) H (760 Torr), palladium on carbon, THF–MeOH, rt; (e)
2
3
2
H O aq., K CO , DMSO, 0°C to rt.
2
2
2
3
Synthesis of 4-(3-pyridiyl)-thiazole and 4-(2-pyridyl) thiazole analogs is described in
Scheme 4. 4-(6-Methoxy-3-pyridiyl)-thiazole 17 was prepared from the intermediate 8a.
Conversion of its methoxy group into a nitrile group afforded compound 20. This
conversion was achieved in three steps; acidic hydrolysis with hydrogen bromide,
chlorination with phosphorus oxychloride, and palladium-catalyzed cyanation. Basic
hydrolysis of the nitrile in 20 afforded the target 4i. O-Methylation of the hydroxypyridine
2
2
1 followed by acylation and bromination at the benzyl position afforded the bromoacetone
4. Introduction of a thiazole ring in 6a yielded 4-(5-methoxy-2-pyridiyl) thiazole 25. Then,
its methoxy moiety was converted to the carboxamide 4j in the same manner as that
described for 4i.
a
Scheme 4. Preparation of pyridyl thiazole analogs 4i–j

a
Reagents and conditions: (a) 6-methoxypyridine-3-boronic acid, Pd (dba) , X-Phos, Cs CO ,
2
3
2
3
DME–water, 80°C; (b) hydrogen bromide, acetic acid, 80 °C; (c) phosphorus oxychloride,
00°C; (d) zinc cyanide, Pd(Ph P) , NMP, 100°C; (e) K CO , H O aq., DMSO, rt; (f)
1
3
4
2
3
2
2
iodomethane, potassium tert-butoxide, THF, 0°C to rt; (g) n-butyl lithium, THF, 0°C; then,
dimethylacetamide; (h) NBS, MeCN, 0°C; (i) 6a, EtOH, 50°C; (j) hydrogen bromide, acetic
acid, reflux; (k) trifluoromethanesulfonic anhydride, pyridine, 0°C; (l) zinc cyanide,
Pd(Ph P) , NMP, 150°C.
3
4
An efficient synthetic route for optimization of the substituents on the benzyl group is
developed in Scheme 5. 2-Formyl-4-methylthiazole was allowed to react with sodium
borohydride to afford the alcohol 29, followed by bromination at the 5-position and Suzuki-
Miyaura coupling reaction to give the ester analog 31. The ester moiety of compound 31

was converted to carboxamide in the same manner described in Scheme 1 to afford
compound 33. Chlorination of 33 provided the key coupling partner 34. Suzuki-Miyaura
coupling reaction of 34 with commercially available arylboronic acids yielded compounds
4
k–p.
a
Scheme 5. Introduction of distal aryl groups in the late stage 4k–p
a
Reagents and conditions: (a) sodium borohydride, MeOH, 0°C; (b) NBS, MeCN, rt; (c)
Pd(Ph P) , Na CO , DME–water, 80°C; (d) NaOH aq., EtOH, reflux; (e) ammonium
3
4
2
3
chloride, EDCI, HOBt, N,N-diisopropylethylamine, DMF, rt; (f) thionyl chloride, 0°C; (g)
arylboronic acid, PdCl (dppf)-CH Cl , Cs CO , THF–water, 150°C, microwave.
2
2
2
2
3
Preparation of the pyrazole analog 4s is described in Scheme 6. The
trifluoromethanesulfonate 38 was prepared via Knoevenagel reaction of the arylacetic acid
3
5 with Meldrum’s acid, followed by cyclization reaction with methylhydrazine and
triflation with triflic anhydride. A subsequent three-step reaction afforded the target
compound 4s as described in thiazole synthesis.
a
Scheme 6. Preparation of pyrazole analog 4s

a
Reagents and conditions: (a) Meldrum's acid, EDCI, N,N-dimethyl-4-aminopyridine, rt;
then, MeOH, reflux; (b) methylhydrazine, MeOH, rt; (c) trifluoromethanesulfonic
anhydride, pyridine, −20°C to rt; (d) ethyl 2-methyl-4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)benzoate, Pd(Ph P) , Na CO , DME, 90°C; (e) NaOH aq., EtOH, 80°C;
3
4
2
3
(
f) WSC, HOBt-NH , DMF, rt.
3
The imidazole analog 4t was synthesized according to Scheme 7. The benzoic acid
analog 41 was converted to the ester 42, and then the regioselective S Ar reaction with 4-
N
iodo-2-methyl-1H-imidazole gave the iodoimidazole analog 43. Introduction of a benzyl
group at the 4-position was achieved via Negishi coupling reaction to yield compound 44.
Using compound 44, the target imidazole analog 4t was prepared in the same manner as
previously described.
a
Scheme 7. Preparation of imidazole analog 4t

a
Reagents and conditions: (a) EtOH, thionyl chloride, 80°C; (b) 4-iodo-2-methyl-1H-
imidazole, sodium hydride, 18-crown-6, NMP, 110°C; (c) 1-(bromomethyl)-3-fluoro-5-
trifluoromethyl)benzene, zinc, 1,2-dibromobutane, chlorotrimethylsilane, THF,
(
palladium(II) acetate, 2,2'-bis-(diphenylphosphino)-1,1'-binaphthyl, THF, 80°C; (d) NaOH
aq., EtOH, rt; (e) EDCI, HOBt-NH , DMF, rt.
3
Preparation of the 1,2,4-triazole analog is described in Scheme 8. S-Methylation of the
thioamide 6c with iodomethane, and the successive treatment with the arylhydrazine 46,
which was prepared from the fluorobenzene 42 and hydrazine, afforded the hydrazone
analog 48. Construction of the 1,2,4-triazole using the hydrazone 48 and orthoester,
followed by conversion of the ester moiety to carboxamide yielded the target 1,2,4-triazole
analog 4u.
a
Scheme 8. Preparation of 1,2,4-triazole analog 4u

a
Reagents and conditions: (a) hydrazine monohydrate, DMSO, 120°C; (b) iodomethane,
acetone, 50°C; (c) EtOH, 0°C; (d) 1,1,1-trimethoxyethane, acetic acid, 100°C; (e) NaOH
aq., EtOH, 80°C; (f) EDCI, HOBt-NH , DMF, rt.
3
3
. BIOLOGICAL RESULTS AND DISCUSSION
We measured the intracellular levels of cAMP in human GPR52-expressing CHO cells at
various concentrations of the compounds. The half maximal effective concentration (EC )
5
0
was defined as the concentration of the compound that results in cAMP production equal to
half of its Emax value, and the cAMP level produced by the reference compound (N-(2-
5
hydroxyethyl)-3-(2-(3-(trifluoromethyl)benzyl)-1-benzofuran-4-yl)benzamide)
was
defined as 100%.

The thiazole derivative 4a was first prepared to validate our design hypothesis that
combined the privileged structure and the azolylbenzamide (3). Compound 4a showed a
significantly higher potency (EC = 220 nM, E = 93%) than compound 3 (EC = 470 nM,
50
max
50
Emax = 56%) as shown in Table 1. Therefore, 4a was selected as a new lead for further
modification. Table 1 shows the modifications of ring A. Introduction of a methyl group on
ring A at the C2 position (4b) was tolerated with an EC value of 73 nM, whereas
5
0
methylation at the C3 position (4c) diminished its GPR52 agonistic activity (EC = 1400
5
0
nM). In general, introduction of a substituent at the ortho-position of a biaryl structure
1
0
induces a dihedral angle at the biaryl . This can be interpreted that the planar biaryl
structure in this azolylbenzamide series was essential for the potency, similar to the amide
structure with S-O interaction in 2. Further exploration of the structure–activity relationship
(
SAR) at the C2 position showed that the ethyl (4d), methoxy (4g), and trifluoromethyl
4h) analogs exhibited similar GPR52 agonistic activities with EC values of 83, 170, and
(
5
0
1
10 nM, respectively. The results demonstrated that the electrostatic potential of the
substituents at the C2 position did not affect the potency. However, the isopropyl derivative
e showed weaker agonistic activity (EC = 370 nM, E = 63%). Introduction of fluorine
4
5
0
max
(
4f) and 2-pyridine (4i) clearly lowered the potency (EC = 380 nM, E = 84% for 4f;
50 max
EC > 1000 nM, E = 11% for 4i at 1µM) compared with the unsubstituted benzene (4a,
5
0
max
EC = 220 nM), which indicated that intramolecular hydrogen bonding between the amide
5
0
and fluorine or pyridine could disrupt the interaction between the ligand and GPR52. The
-pyridyl derivative 4j, exhibited an EC value of 400 nM with 70% efficacy, which was
3
5
0
weaker than that of the benzene derivative (4a; EC = 220 nM, E = 93%). Thus, it was
5
0
max

concluded that the 2-methyl-benzamide structure is functionally suitable for GPR52
agonists.
Table 1. Effects of substituents at the ring A on GPR52 agonistic activity.
a
human GPR52
1
Compound
ring A
Ph
R
EC (nM)
Emax (%)
50
4
4
4
4
4
4
4
4
4
a
b
c
d
e
f
H
220 (180–270)
73 (56–96)
93 (89–97)
86 (82–91)
87 (83–91)
84 (79–89)
63 (54–72)
84 (82–86)
88 (85–91)
92 (91–94)
Ph
2-Me
3-Me
2-Et
2-iPr
2-F
Ph
1400 (1200–1700)
83 (64–107)
Ph
Ph
370 (240–570)
380 (340–430)
170 (140–210)
110 (100–120)
>1000
Ph
g
h
i
Ph
2-OMe
2-CF3
H
Ph
2-Py
3-Py
11%@1µM
4
j
H
400 (350–450)
70 (69–72)
a
EC values are derived from the mean curves of the experiments (n = 2–4). Numbers in
parentheses represent the 95% confidence interval.
5
0
According to the previously described SAR of compounds 1 and 2, substituents on ring
C affected the agonistic activity, in particular, substituents at the C3 position had great
impact on the potency. Detailed study of the effects of C3-substituents in compound 4 was

conducted for further improvement of the GPR52 agonistic activity (Table 2). The methoxy
4k), N,N-dimethylamine (4l), fluorine (4m), and chlorine (4n) derivatives showed high
potency with EC values of 160, 690, 260, and 130 nM, respectively. Therefore, it was
(
50
concluded that the electronic nature of substituents at the C3 position did not affect the
potency. Chlorine scan, which is a common drug design strategy, was then performed to
identify the most promising spot. The GPR52 agonistic activity and efficacy of the ortho-
substituted (4o) and para-substituted (4p) derivatives (EC50 = 3900 and 210 nM,
respectively; Emax = 51% and 60%, respectively) were lower than those of the meta-
substituted derivative (4n; EC = 130 nM, E = 75%). Thus, meta-substitution was
50
max
considered advantageous with respect to both the potency and efficacy. These results led us
to introduce multiple substituents at the meta-position, such as 3-chrolo-5-fluoro (4q) and
3
-fluoro-5-trifluoromethyl (4r). Compounds 4q and 4r were found to be potent GPR52
agonists with EC values of 25 and 11 nM, respectively. Thus, substituents at the meta-
5
0
position on ring C greatly affected the functional GPR52 activity in a similar manner to that
previously reported with compounds 1 and 2.
Table 2. Effects of substituents at the ring C on GPR52 agonistic activity.
a
human GPR52
2
Compound
R
EC (nM)
Emax (%)
5
0

4
4
4
4
4
4
4
k
l
3-OMe
3-NMe2
3-F
160 (120–220)
690 (550–860)
260 (180–370)
130 (95–190)
3900 (1700–8700)
210 (150–290)
25 (18–35)
85 (79–90)
101 (95–107)
75 (70–81)
75 (70–80)
51 (42–60)
60 (55–64)
88 (84–92)
93 (90–96)
m
n
o
p
q
r
3-Cl
2-Cl
4-Cl
3-Cl-5-F
3-F-5-CF3
4
11 (9.2–14)
a
EC values are derived from the mean curves of the experiments (n = 2–4). Numbers in
parentheses represent the 95% confidence interval.
5
0
Modification of the ring B was then conducted to explore the effect of other azoles. The
2
position and number of nitrogen atoms with sp -hybridization affected the electrostatics of
the ring, which, in turn, affected the basicity, lipophilicity, physicochemical properties, and
PK profiles of a rat cassette-dosing test. The GPR52 agonistic activity and PK profiles of
4
r–u are highlighted in Table 3. The pyrazole analog 4s showed a GPR52 agonistic activity
with an EC value of 23 nM, which was comparable to that of the thiazole analog 4r.
5
0
Furthermore, the imidazole analog 4t and the 1,2,4-triazole analog 4u exhibited an
agonistic activity at a double-digit nanomolar concentrations (EC = 59 and 75 nM,
50
respectively). A previous series of the thiazole analog 2 considered only the thiazole ring as
the core motif for potent GPR52 agonistic activity. However, this series of the
azolylbenzamide (4) derivatives accepted various five-membered azoles as the core motif.
These results can validate the design concept of 4, which featured the comparability
between the benzene in 4 and the amide with S-O interaction in 2.
Table 3. GPR52 agonistic activity and PK parameters in rat of azolyl benzamide 4r–u.

a
human GPR52
b,c
b,d
b,e
Compo ring B
und
EC50
Emax
F
AUCpo
Cltotal
log
f
D
(
nM)
(%)
(%) (ng*h/mL) (mL/h/kg)
4
4
4
r
11
93 (90–96) 14.0 60.5
2369
1975
1129
1409
5.3
4.2
4.7
3.8
(
9.2–14)
s
23
123 (118– 30.5 155.4
29)
1
(
19–29)
t
59
107 (103– 28.4 267.9
10)
1
(
49–71)
4
u
75
122 (115– 53.8 383.9
30)
1
(
60–95)
a
EC values are derived from the mean curves of the experiments (n = 2–4). Numbers in
5
0
b
parentheses represent the 95% confidence interval. Cassette-dosing. Male SD rats (n = 3).
c
Dose: i.v. at 0.1 mg/kg; p.o. at 1 mg/kg. Bioavailability. Area under the curve from 0 to 8
d
e
f
11
hours. Clearance. logD at pH 7.4 was calculated using ACD Labs version 12 .

Among the compounds shown in Table 3, the 1,2,4-triazole analog 4u possessed better
bioavailability (F = 53.8%) than compounds 4r–t. Furthermore, compound 4u showed the
highest value of area under the curve (AUC) (383.9 ng*h/mL) and a moderate total
po
clearance (Cl = 1409 mL/h/kg). Regarding the modification of ring B, reduction of
lipophilicity was closely associated with lower clearance rate, resulting in improvement of
AUC and bioavailability. Moreover, compound 4u showed good metabolic stability
po
against microsomes of several species (metabolic clearance: 17 µL/min/mg for humans, 9
µL/min/mg for rats, 24 µL/min/mg for mice). Therefore, we concluded that low
lipophilicity resulted in high bioavailability owing to slow clearance as the result of high
metabolic stability. In addition, compound 4u did not inhibited off-targets such as receptors
of dopamine D , D , 5-HT , AMPA, NMDA, and phosphodiesterases with the values of
1
2L
2
12
under 50% inhibition at 10 µM concentration . Based on these results, compound 4u,
which showed potent GPR52 activity, high metabolic stability in several species, and
excellent PK in rats, was selected for further biological in vivo evaluation in mice.
Detailed PK profile of 1,2,4-triazole derivative 4u (Table 4) was determined in ICR mice
using both plasma and brain tissues. The concentrations of 4u in plasma and brain were
measured after an oral administration of 10 or 30 mg/kg (n = 3 in each). Compound 4u
showed good permeability through the blood-brain barrier (BBB) with brain/plasma ratio of
1
.53–1.68. It successfully demonstrates that AUC0-24h values were increased with a dose
escalation in both plasma and brain.
Table 4. PK parameters of compound 4u in mice.
a

Dosing
10 mg/kg
Plasma
30 mg/kg
Plasma
Sample tissue
Brain
1.018
Brain
2.330
Cmax
0.688
1.209
(µg/mL or µg/g)
Tmax (h)
0.50
0.50
1.00
1.00
AUC0-24h
2.433
3.731
6.200
10.414
(µg*h/mL or µg*h/g)
b
Brain/Plasma ratio
1.53
1.68
3.54
MRT (h)
2.29
2.06
3.25
a
b
ICR mice (n = 3 in each). The values were calculated with AUC0-24h(brain)/AUC0-
24h(
plasma).
In addition to its excellent PK profile in rodents, it was revealed that compound 4u
possessed mice GPR52 potency (EC = 71 nM, E = 114%) with a plasma protein binding
5
0
max
value of 80.1% in mice at 1 µM concentration. Therefore, the effects of compound 4u were
investigated in animal models of psychiatric disorders. The dopamine agonist,
methamphetamine (MAP), induces some effects that mimic the positive symptoms of
schizophrenia. In a MAP-induced (2 mg/kg, s.c) hyperlocomotion activity assay using male
ICR mice, compound 4u showed efficacy in the prevention of the hyperlomotion after oral
administration at a dose of 10 mg/kg, as shown in Figure 4. Compound 4u did not show
cataleptogenic effects (unpublished data).

Figure 4. Suppression of MAP-induced hyperactivity by compound 4u in mice. (Male ICR
#
mice, mean ± SEM, ***p < 0.01 by t-test, p < 0.025 by Williams test).
4
. CONCLUSION
Azolylbenzamide 4a–u was designed from the high-throughput screening hit 3, based on
the previously reported GPR52 agonist motif, particularly the 2-benzyl-4-methylthiazole. In
the SAR study of 4, it was clear that (1) the ring A benzene motif in 4 could replace the
amide motif in 2; and (2) a planar biaryl system exhibited higher potency over the twisted
biaryl structure based on the analysis of dihedral angles. Moreover, introduction of a
methyl group at the C2 position of ring A and halogen atoms or trifluoromethyl group at

the meta-position of ring C enhanced the GPR52 agonistic activity. Additionally, the ring B
tolerated various five-membered azoles, resulting in the identification of potent GPR52
agonists, such as the thiazole (4r), pyrazole (4s), imidazole (4t), and 1,2,4-triazole (4u)
analogs. Among them, the 1,2,4-triazole analog 4u showed a good PK profile and BBB
penetration. Moreover, MAP-induced hyperlocomotion was significantly suppressed by
oral administration of 4u at a dose of 10 mg/kg. Detailed biological evaluation will be
reported in the near future.
5
5
.
EXPERIMENTAL SECTION
.1. CHEMISTRY
1
The proton nuclear magnetic resonance ( H-NMR) spectra were determined on a Bruker
AVANCE II (300 MHz) or Varian INOVA-400 (400 MHz) instruments. Chemical shifts
1
for H-NMR were reported in parts per million (ppm) downfield from tetramethylsilane (δ)
as the internal standard in deuterated solvent and coupling constants (J) are in Hertz (Hz).
The following abbreviations are used for spin multiplicity: s = singlet, d = doublet, t =
triplet, q = quartet, m = multiplet and br = broad. Elemental analyses were carried out by
Takeda Analytical Research Laboratories, Ltd. Analytical thin layer chromatography (TLC)
was performed on silica gel 60 F plates (Merck) or NH TLC plate (Fuji Silysia Chemical
254
®
Ltd.). Column chromatography was carried out on a silica gel column (Chromatorex NH-
DM1020, 100-200 mesh, Fuji Silysia chemical) or on Purif-Pack (SI • 60 µM or NH • 60
µM, Fuji Silysia Chemical, Ltd.). Liquid chromatography-mass spectrometry (LC/MS)
analysis was performed on the following methods; (A) an Agilent 1200, equipped with a L-
column2 ODS (3.0 x50 mm I.D., 3 µm-particle size, CERI, Japan), eluting with 5 mM

4
ammonium acetate (AcONH ) in ultrapure water/acetonitrile=90/10 (Mobile phase A), and
5
mM AcONH in ultrapure water/acetonitrile=10/90 (Mobile phase B), using the
4
following elution gradient of 5% B to 90% B over 0.9 min followed by 90% B isocratic
over 1.1 min at a flow rate of 1.5 mL/min (detection at 220 nm or 254 nm). MS spectra
were recorded using an Agilent 6130 with electrospray ionization. (B) a Shimadzu LC-
2
0AD, equipped with a L-column2 ODS (3.0 x50 mm I.D., 3 µm-particle size, CERI,
Japan), eluting with 5 mM AcONH in ultrapure water/acetonitrile = 90/10 (Mobile phase
A), and 5 mM AcONH in ultrapure water/acetonitrile = 10/90 (Mobile phase B), using the
4
4
following elution gradient of 5% B to 90% B over 0.9 min followed by 90% B isocratic
over 1.1 min at a flow rate of 1.5 mL/min (detection at 220 nm or 254 nm). MS spectra
were recorded using a Shimadzu LCMS-2020 with electrospray ionization. (C) a Shimadzu
LC-20AD, equipped with a L-column2 ODS (3.0 x50 mm I.D., 3 µm-particle size, CERI,
Japan), eluting with 0.05% TFA in ultrapure water (Mobile phase A), and 0.05% TFA in
acetonitrile (Mobile phase B), using the following elution gradient of 5% B to 90% B over
0
.9 min followed by 90% B isocratic over 1.1 min at a flow rate of 1.5 mL/min (detection
at 220 nm). MS spectra were recorded using a Shimadzu LCMS-2020 with electrospray
ionization. (D) a Waters 2795, equipped with a L-column2 ODS (3.0 x50 mm I.D., 3µm-
particle size, CERI, Japan), eluting with 0.05% TFA in ultrapure water (Mobile phase A),
and 0.05% TFA in acetonitrile (Mobile phase B), using the following elution gradient of
5
% B to 90% B over 2 min followed by 90% B isocratic over 1.5 min at a flow rate of 1.2
mL/min (detection at 220 nm or 254 nm). MS spectra were recorded using a Waters
ZQ2000 with electrospray ionization. Preparative LC was performed on a Waters 2525

separations module (L-column2 ODS (20 x150 mm I.D., CERI, Japan); 0.1% TFA in
distilled water/acetonitrile gradient; MS spectra were recorded using a Waters ZQ2000 with
electrospray ionization. The purities of all compounds tested in biological systems were
assessed as being >95% using analytical high-performance liquid chromatography (HPLC).
Purity data were collected by a HPLC with Corona CAD (Charged Aerosol Detector) or
photo diode array detector. The column was a Capcell Pak C18AQ (50 mm x 3.0 mm I.D.,
Shiseido, Japan) or L-column 2 ODS (30 mm x 2.0 mm I.D., CERI, Japan) with a
temperature of 50 °C and a flow rate of 0.5 mL/min. Mobile phase A and B under a neutral
4
condition were a mixture of 50 mmol/L AcONH , water and acetonitrile (1:8:1, v/v/v) and a
mixture of 50 mmol/L AcONH and acetonitrile (1:9, v/v), respectively. The ratio of mobile
4
phase B was increased linearly from 5% to 95% over 3 min, 95% over the next 1 min.
Mobile phase A and B under an acidic condition were a mixture of 0.2% formic acid in 10
mmol/L ammonium formate and 0.2% formic acid in acetonitrile, respectively. The ratio of
mobile phase B was increased linearly from 14% to 86% over 3 min, 86% over the next 1
min. All commercially available solvents and reagents were used without further
purification. Yields were not optimized. Abbreviations: B pin , bis(pinacolato) diboron;
2
2
Pd (dba) ,
tris(dibenzylideneacetone)
bis(triphenylphosphine) palladium(II);
bis(diphenylphosphino)ferrocene] palladium(II)
dicyclohexylphosphino-2',4',6'-triisopropylbiphenyl; Na CO , sodium carbonate; NaHCO ,
dipalladium(0);
PdCl (Ph P) ,
dichloro
2
2
2
3
2
PdCl (dppf)-CH Cl ,
dichloro
[1,1’-
2
2
2
CH Cl
adduct;
X-Phos,
2-
2
2
2
3
3
sodium hydrogen carbonate; NaOH, sodium hydroxide; K CO , potassium carbonate;
2
3
Cs CO , cesium carbonate; MgSO , magnesium sulfate; NH Cl, ammonium chloride; EDCI,
2
3
4
4

1
-(3-dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride;
HOBt,
1-
hydroxybenzotriazole; NBS, N-bromosuccinimide; H O , hydrogen peroxide; HCl,
2
2
hydrochloric acid; DME, 1,2-dimethoxyethane; DMF, N,N-dimethylformamide; NMP, N-
methylpyrrolidone; EtOAc, ethyl acetate; MeOH, methyl alcohol; EtOH, ethyl alcohol;
THF, tetrahydrofuran.
5
.1.1. 2-(3-(Trifluoromethyl)phenyl)ethanethioamide (6a)
A mixture of (3-(trifluoromethyl)phenyl)-acetonitrile (45.2 g, 244 mmol) and diethyl
dithiophosphate (50.0 g, 269 mmol) in 4 N HCl solution in EtOAc (500 mL, 2000 mmol)
was stirred at room temperature overnight. The mixture was washed with water, 1 N
aqueous NaOH solution, and brine, dried over Na SO , and concentrated in vacuo. The
2
4
residue was triturated with hexane with ice-bath cooling. The precipitate was collected by
1
filtration followed by washing with hexane to give 35.0 g (65%) of 6a as colorless solid. H
NMR (300 MHz, CDCl ) δ4.14 (2H,s), 6.66 (1H, brs), 7.38–7.88 (5H, m).
3
5
.1.2. 2-(3-Chloro-5-fluorophenyl)ethanethioamide (6b)
Compound 6b was obtained as pale yellow solid in 68% yield by a method similar to that
1
described for 6a. H NMR (300 MHz, CDCl ) δ4 .. 03 (2H, s), 6.78 (1H, brs), 6.96 (1H, dd, J
3
=
8.9, 1.7 Hz), 7.03–7.14 (2H, m), 7.81 (1H, brs).
5
.1.3. 2-(3-Fluoro-5-(trifluoromethyl)phenyl)ethanethioamide (6c)
Compound 6c was obtained as colorless solid in 83% yield by a method similar to that
1
described for 6a. H NMR (300 MHz, CDCl ) δ4.10 (2H, s), 6.69 (1H, brs), 7.21–7.41 (3H,
3
m), 7.56 (1H, brs).

5
.1.4. 4-Methyl-2-(3-(trifluoromethyl)benzyl)-1,3-thiazole (7a)
A mixture of 6a (4.38 g, 20 mmol) and 80% bromoacetone (2.3 mL, 22 mmol) in EtOH (20
mL) was refluxed for 2 h. After cooling, the mixture was concentrated in vacuo. The
mixture was diluted with EtOAc, washed with saturated aqueous NaHCO solution and
3
brine, dried over Na SO and concentrated in vacuo. The residue was purified by column
2
4
chromatography (silica gel, eluted with EtOAc in hexane) to give 4.97 g (97%) of 7a as
1
colorless oil. H NMR (300MHz, CDCl ) δ2.44 (3H, s), 4.35 (2H, s), 6.77 (1H, s), 7.33–
3
7
5
.67 (4H, m).
.1.5. 2-(3-Chloro-5-fluorobenzyl)-4-methyl-1,3-thiazole (7b).
Compound 7b was obtained as pale brown oil in quantitative yield by a method similar to
1
that described for 7a. H NMR (300 MHz, CDCl ) δ2.44 (3H, s), 4.25 (2H, s), 6.78 (1H, s),
3
6
5
.89–7.03 (2H, m), 7.10 (1H, s).
.1.6. 2-(3-Fluoro-5-(trifluoromethyl)benzyl)-4-methyl-1,3-thiazole (7c)
Compound 7c was obtained as pale yellow oil in 95% yield by a method similar to that
1
described for 7a. H NMR (300 MHz, CDCl ) δ2.45 (3H, d, J = 0.8 Hz), 4.34 (2H, s), 6.80
3
(
1H, q, J = 1.1 Hz), 7.22 (2H, d, J = 9.0 Hz), 7.38 (1H, s).
.1.7. 5-Bromo-4-methyl-2-(3-(trifluoromethyl)benzyl)-1,3-thiazole (8a)
5
To a solution of 7a (1.29 g, 5.01 mmol) in MeCN (10 mL) was added NBS (0.979 g, 5.50
mmol) with several portions at 0 °C. The mixture was stirred at room temperature overnight.
The mixture was poured into saturated aqueous NaHCO solution and extracted with
3
EtOAc. The organic layer was separated, washed with brine, dried over Na SO and
2
4
concentrated in vacuo. The residue was purified by column chromatography (silica gel,

1
eluted with 1%–10% EtOAc in hexane) to give 0.907 g (54%) of 8a as yellow oil. H NMR
(
300 MHz, CDCl ) δ2.38 (3H, s), 4.28 (2H, s), 7.42–7.60 (4H, m).
3
5
.1.8. 5-Bromo-2-(3-chloro-5-fluorobenzyl)-4-methyl-1,3-thiazole (8b)
Compound 8b was obtained as pale oil in 82% yield by a method similar to that
1
described for 8a. H NMR (300 MHz, CDCl ) δ2.38 (3H, s), 4.18 (2H, s), 6.88–6.94 (1H,
3
m), 6.98–7.04 (1H, m), 7.08 (1H, s).
5
.1.9. 5-Bromo-2-(3-fluoro-5-(trifluoromethyl)benzyl)-4-methyl-1,3-thiazole (8c)
Compound 8c was obtained as pale brown oil in 99% yield by a method similar to that
1
described for 8a. H NMR (300 MHz, CDCl ) δ2.39 (3H, s), 4.28 (2H, s), 7.16–7.38 (3H,
3
m).
5
.1.10. Ethyl 4-(4-methyl-2-(3-(trifluoromethyl)benzyl)-1,3-thiazol-5-yl)benzoate (10a)
A mixture of 8a (1.35 g, 4.00 mmol), (4-(ethoxy-carbonyl)phenyl)boronic acid (9a; 853
mg, 4.40 mmol), Pd(Ph P) (139 mg, 0.12 mmol), 2 M aqueous Na CO solution (4.4 mL,
3
4
2
3
8
.8 mmol), and DME (18 mL) was stirred at 80°C overnight. The mixture was diluted with
EtOAc, washed with brine, dried over Na SO , and concentrated in vacuo. The residue was
2
4
purified by column chromatography (silica gel, eluted with 5%–25% EtOAc in hexane) to
1
give 1.53 g (94%) of 10a as pale yellow oil. H NMR (300 MHz, CDCl ) δ1.40 (3H, t, J =
3
7
.2 Hz), 2.51 (3H, s), 4.32–4.44 (4H, m), 7.41–7.65 (6H, m), 8.02–8.10 (2H, m).
.1.11. Methyl 3-methyl-4-(4-methyl-2-(3-(trifluoromethyl)benzyl)thiazol-5-yl)benzoate
10b)
A mixture of 8a (338 mg, 1.00 mmol), methyl 3-methyl-4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)benzoate (9b; 331 mg, 1.2 mmol), Pd(Ph P) (35 mg, 0.030 mmol), 2 M
5
(
3
4

aqueous Na CO solution (1.2 mL, 2.4 mmol) in DME (3.6 mL) was stirred at 65 °C
2
3
overnight. The mixture was diluted with EtOAc, washed with brine, dried over Na SO , and
2
4
concentrated in vacuo. The residue was purified by column chromatography (silica gel,
1
eluted with 5%–25% EtOAc in hexane) to give 280 mg (69%) of 10b as colorless oil. H
NMR (300 MHz, CDCl ) δ2.22 (3 HH , s), 2.2 55 (3H, s) ,, 3.93 (3 HH , s), 4.3 77 (2H, s) ,, 7.29 (1H, d,
3
J = 7.9 Hz), 7.44–7.65 (4H, m), 7.86 (1H, dd, J = 7.9, 1.9 Hz), 7.93–7.96 (1H, m).
5
.1.12. Methyl
yl)benzoate (10c)
A mixture of 8a (335 mg, 1.00 mmol), (3-fluoro-4-(methoxycarbonyl)phenyl)boronic
acid (9c; 297 mg, 1.50 mmol), Pd (dba) (46 mg, 0.050 mmol), X-Phos (48 mg, 0.10 mmol),
2-fluoro-4-(4-methyl-2-(3-(trifluoromethyl)benzyl)-1,3-thiazol-5-
2
2
Cs CO (652 mg, 2.0 mmol), DME (2 mL), and water (0.5 mL) was stirred at 50°C for 3 h.
2
3
The mixture was diluted with EtOAc, washed with brine, dried over Na SO , and
2
4
concentrated in vacuo. The residue was purified by column chromatography (silica gel,
1
eluted with 3%–30% EtOAc in hexane) to give 258 mg (63%) of 10c as colorless solid. H
NMR (300 MHz, CDCl ) δ 2.53 (3H, s), 3.94 (3H, s), 4.35 (2H, s), 7.15–7.25 (2H, m),
3
7
.42–7.65 (4H, m), 7.96 (1H, t, J = 7.9 Hz).
.1.13. Methyl 2-methoxy-4-(4-methyl-2-(3-(trifluoromethyl)benzyl)-1,3-thiazol-5-
yl)benzoate (10d)
A mixture of 8a (335 mg, 1.0 mmol), (3-methoxy-4-(methoxycarbonyl)phenyl)boronic
acid (9d; 315 mg, 1.5 mmol), Pd (dba) (46 mg, 0.05 mmol), X-Phos (48 mg, 0.10 mmol),
5
2
2
Cs CO (652 mg, 2.0 mmol) in DME (2 mL) and water (0.5 mL) was stirred at 50°C for 3 h.
2
3
The mixture was diluted with EtOAc, washed with brine, dried over Na SO , and
2
4