Fused bicyclic heteroarylpiperazine-substituted l-prolylthiazolidines as highly potent DPP-4 inhibitors lacking the electrophilic nitrile group

Article abstract of DOI:10.1016/j.bmc.2012.06.033

Hypoglycemic agents with a mechanism of depeptidyl peptidase IV (DPP-4) inhibition are suitable for once daily oral dosing. It is difficult to strike a balance between inhibitory activity and duration of action in plasma for inhibitors bearing an electrophilic nitrile group. We explored fused bicyclic heteroarylpiperazine substituted at the γ-position of the proline structure in the investigation of l-prolylthiazolidines lacking the electrophilic nitrile. Among them, 2-trifluoroquinolyl compound 8g is the most potent, long-lasting DPP-4 inhibitor (IC₅₀ = 0.37 nmol/L) with high selectivity against other related peptidases. X-ray crystal structure determination of 8g indicates that CH-π interactions generated between the quinolyl ring and the guanidinyl group of Arg358 enhances the DPP-4 inhibitory activity and selectivity.

Full text of DOI:10.1016/j.bmc.2012.06.033

Bioorganic & Medicinal Chemistry 20 (2012) 5033–5041
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Fused bicyclic heteroarylpiperazine-substituted L-prolylthiazolidines
as highly potent DPP-4 inhibitors lacking the electrophilic nitrile group
Tomohiro Yoshida a, , Fumihiko Akahoshi a, , Hiroshi Sakashita a,à, Shuji Sonda a,§, Masahiro Takeuchi a,
⇑
Yoshihito Tanaka a, Mika Nabeno a, Hiroyuki Kishida b, Ikuko Miyaguchi b, Yoshiharu Hayashi a,
a Research Division, Mitsubishi Tanabe Pharma Corporation, 2-2-50, Kawagishi, Toda-shi, Saitama 335-8505, Japan
b R&D Strategy Dept., Corporate Strategy Division, Mitsubishi Chemical Corporation, 1000, Kamoshida-cho, Aoba-ku, Yokohama 227-0033, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Hypoglycemic agents with a mechanism of depeptidyl peptidase IV (DPP-4) inhibition are suitable for
once daily oral dosing. It is difficult to strike a balance between inhibitory activity and duration of action
in plasma for inhibitors bearing an electrophilic nitrile group. We explored fused bicyclic heteroarylpip-
erazine substituted at the c-position of the proline structure in the investigation of L-prolylthiazolidines
lacking the electrophilic nitrile. Among them, 2-trifluoroquinolyl compound 8g is the most potent, long-
Received 28 April 2012
Revised 7 June 2012
Accepted 8 June 2012
Available online 5 July 2012
lasting DPP-4 inhibitor (IC50 = 0.37 nmol/L) with high selectivity against other related peptidases. X-ray
Keywords:
DPP-4 inhibitor
Thiazolidine
crystal structure determination of 8g indicates that CH-
p interactions generated between the quinolyl
ring and the guanidinyl group of Arg358 enhances the DPP-4 inhibitory activity and selectivity.
Ó 2012 Elsevier Ltd. All rights reserved.
Bicyclic heteroarylpiperazine
1. Introduction
serine protease recognizing an amino acid sequence having proline
or alanine at the N-terminal penultimate position to produce a
The number of patients with type 2 diabetes is increasing rap-
idly worldwide.1 Diabetics may suffer debilitating cardiovascular,
eye, kidney, and nerve damage and are at risk of premature hand-
icap and death due to these and other diabetic complications,
which are the result of glucose toxicity caused by hyperglycemia.
Current treatment strategies include reducing insulin resistance,
supplementing the insulin deficiency with exogenous insulin,
enhancing endogenous insulin secretion, reducing hepatic glucose
output, and limiting glucose absorption.2 Among them, insulin
secretagogues such as glinides (e.g., repaglinide), which are widely
used, need to be taken before each meal for avoiding a risk of hypo-
glycemia. Therefore the antihyperglycemic agent which is orally
available in a once-daily regimen without causing hypoglycemia
had been much awaited in medical settings.
dipeptide and the inactive GLP-1.3 The DPP-4 inhibition increases
the plasma concentration of active GLP-1, causes the secretion of
insulin in response to increased blood glucose level;4 therefore,
DPP-4 inhibitors are expected to be safer and once daily agents.5
Several DPP-4 inhibitors have been already approved, including
sitagliptin (1, first in class), vildagliptin (2), alogliptin (3) and sax-
agliptin (4) (Fig. 1).6 Several other candidates have been reported to
be in an advanced stage in clinical trials for type 2 diabetes.5b
Many known DPP-4 inhibitors have the P1–P2 fragment7 and
some of them possess the electrophilic trap such as a nitrile or
boronic acid group on the P1 pyrrolidine to form a covalent bond
with Ser630 of the catalytic triad in the active site.8 These
F
OH
F
NH2 O
An incretin hormone glucagon-like peptide-1 (GLP-1) stimu-
lates insulin biosynthesis and secretion in response to meal inges-
tion, inhibits glucagon secretion, and promotes proliferation of
pancreatic b cells. Active GLP-1 (GLP-1[7-36]amide) is rapidly
degraded in vivo by depeptidyl peptidase IV (DPP-4), which is a
N
N
N
N
N
H
F
N
CN
O
CF3
sitagliptin (1)
vildagliptin (2)
⇑
Corresponding author.
HO
O
CN
NH2
E-mail address: akahoshi.fumihiko@mk.mt-pharma.co.jp (F. Akahoshi).
Present address: Development Division, Mitsubishi Tanabe Pharma Corporation,
17-10, Nihonbashi-Koamicho, Chuo-ku, Tokyo 103-8405, Japan.
H3C
O
N
N
N
H2N
CN
O
N
à
Present address: Tanabe R&D Service Corporation, 1000, Kamoshida-cho, Aoba-
ku, Yokohama 227-0033, Japan.
alogliptin (3)
saxagliptin (4)
§
Present address: CMC Division, Mitsubishi Tanabe Pharma Corporation, 3-16-89,
Kashima, Yodogawa-ku, Osaka-shi, Osaka 532-8505, Japan.
Figure 1. DPP-4 inhibitors.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.06.033

5034
T. Yoshida et al. / Bioorg. Med. Chem. 20 (2012) 5033–5041
HO
compounds were chemically unstable due to intramolecular cycli-
zation between the electrophilic nitrile and the amine of the P2
part,9 and unable to be expected to control plasma glucose all
day long for once-daily treatment. In this class, saxagliptin (4) im-
proved chemical stability by introduction of a cis-4,5-methano
bridge to the prolinenitrile.10 On the other hand, we started re-
search on novel DPP-4 inhibitors without the electrophilic nitrile
moiety to increase chemical stability and focused on the substitu-
O
RR'N
c, d
a, b
S
S
OH
N
N
N
N
PG
N
H
O
PG
O
O
6a: PG = Boc
6b: PG = Cbz
7a: PG = Boc
7b: PG = Cbz
8, 9a, 9d-g
Y
e, f
NH
ent at the c-position of the proline moiety of the prolylthiazolidine
N
core structure to increase further affinity with the S2 subsite of
N
N
HN
DPP-4.11 We previously reported that a program of optimizing
Y
N
c-substituents by introducing 4-arylpiperazine resulted in highly
g, h
S
S
potent and long-lasting inhibitors.12 The SAR study revealed that
the introduction of an electron-deficient 4-arylpiperazine scaffold
was preferable to increase the inhibitory activity. In particular,
substituted pyridylpiperazine compound 5a proved exceptionally
potent in vitro despite the lack of an electrophilic nitrile group
( Fig. 2). Another important feature of DPP-4 inhibitors is their
selectivity against other related prolyl peptidases such as DPP-8
and DPP-9.13 DPP-8 and DPP-9 inhibition have been reported to
be associated with multiorgan toxicities and profound immuno-
toxicity in rats and dogs,14 and DPP-4 selectivity is one of the key
issues for clinical use. Although phenyl analog 5b showed potent
DPP-4 inhibitory activity, it possessed moderate selectivity against
DPP-8 and DPP-9 as described in a later section. We anticipated
that potent, selective and long-lasting DPP-4 inhibitors would be
NH
+
N
N
N
N
Cbz
N
Cl
H
O
10
O
11a: Y = Cl
11b: Y = CN
9b: Y = Cl
9c: Y = CN
Scheme 1. Reagents and conditions: (a) Thiazolidine, HOBt, EDC, DMF; (b) DMSO,
SO3-pyridine, Et3N; (c) RR0NH, NaBH(OAc)3, AcOH, 1,2-dichloroethane; (d) HCl,
AcOEt; (e) N-Boc-piperazine, NaBH(OAc)3, AcOH, 1,2-dichloroethane; (f) THF,
CH2Cl2; (g) iPr2NEt, N-methyl-2-pyrrolidone, 100 °C; (h) 30% HBr–AcOH or TFA,
thioansole.
Table 1
designed by the optimization of the L-prolylthiazolidine moiety
Inhibition of quinoline and isoquinoline derivatives
having extensive interactions with the S2 subsite.
Ar
N
Here we addressed the introduction of a sterically more bulky
aromatic ring such as a fused bicyclic heteroaryl group on the
piperazine moiety to interact the extensive subsite and discovered
a highly potent, selective and long-lasting DPP-4 inhibitor with a
favorable profile as a potential once-daily therapeutic agent for
type 2 diabetes.
N
S
N
N
H
O
Compound
Ar
DPP-4 inhibition, IC50 (nmol/L)
Human
Rat
2. Chemistry
O2N
N
5a
8a
0.92
1.1
The synthesis of a series of fused bicyclic heteroarylpiperazine
compounds is shown in Scheme 1. The key intermediate 7a was
prepared by condensation with N-Boc-trans-4-hydroxyproline
(6a) and thiazolidine, followed by oxidation with dimethyl sulfox-
ide and sulfur trioxide.15 A rapid and efficient synthesis of 8 and 9
was performed using reductive amination of the key intermediate
ketone 7a with a fused bicyclic heteroarylpiperazine followed by
removal of the Boc group, as previously reported.11 This reductive
amination afforded only cis-isomers. Functionalized benzimidazol-
ylpiperazine derivatives 9b and 9c were synthesized through
nucleophilic substitution of piperazine compound 10 and the cor-
responding 2-chloroimidazole 11.
N
0.61
0.73
0.83
0.79
Cl
N
N
8b
8c
NC
0.51
0.56
8d
8e
2.2
3.0
1.0
N
N
3. Results and discussion
0.95
A series of quinoline and isoquinoline derivatives was evaluated
for human and rat DPP-4 inhibitory activity in vitro and these re-
CH3
CF3
N
N
8f
0.56
0.37
1.5
O2N
Ar
X
N
N
8g
0.59
N
N
γ
S
S
N
N
N
N
N
H
N
H
8h
8i
0.61
1.6
0.62
1.3
O
O
Cl
5a: X=N; IC50 = 0.92 nmol/L
5b: X = CH; IC50 = 1.6 nmol/L
Ar: fused bicyclic heteroaryl
F3C
Figure 2. (S)-c-Substituted L-prolythiazolidines as DPP-4 inhibitors.

T. Yoshida et al. / Bioorg. Med. Chem. 20 (2012) 5033–5041
5035
Table 2
sults are listed in Table 1. The unsubstituted 1-isoquinolyl com-
Inhibition of benzimidazole, benzoxazole and benzothiazole derivatives
pound 8a had a little more potent activity than the previously re-
ported 5-nitro-2-pyridyl compound 5a.12 We already reported that
introduction of a nitrile or nitro group on the phenyl or pyridyl ring
led to an increase in activity.12 Introduction of a chloro or nitrile
group at the 4-position of isoquinoline resulted in compounds 8b
and 8c with similar activities to compounds 8a, suggesting that
no further increase of activity would be expected by the substitu-
tion. 4-Quinolyl compound 8e was more suitable than 2-quinolyl
analog 8d. The substituent effect of 4-quinolyl compounds was
investigated to explore the extension of the S2 subsite and the
introduction of a trifluoromethyl group led to the most potent
compound 8g among the reported DPP-4 inhibitors possessing
the prolylthiazolidine core structure.11,12 The introduction of a
chlorine atom at the 7-position of 4-quinolyl maintained the activ-
ity (8h), whereas the introduction of a bulky group such as trifluo-
romethyl decreased the activity (8i). These results indicate that the
S2 subsite of DPP-4 has extensive and shallow space and is respon-
sible for increasing the activity.
Ar
N
N
S
N
N
H
O
Compound
Ar
DPP-4 inhibition, IC50 (nmol/L)
Human
Rat
N
9a
9b
9c
0.80
0.59
0.39
1.1
N
H
Cl
N
1.03
0.61
N
H
NC
N
N
H
N
O
X-ray crystal structure determination shows that 8g made
many interactions with the active site of DPP-4 (Fig. 3). The thiazol-
idine moiety fully occupies the S1 hydrophobic subsite. The sec-
ondary amino group of the proline moiety forms salt bridges to
Glu205 and Glu206, and the carbonyl oxygen forms a hydrogen
bond with Asn710. The quinolyl ring is stacked with the side chain
of Phe357 in the S2 subsite. The trifluoromethyl group on the qui-
nolyl ring also interacted with Tyr58516 and this interaction
explained that compound 8f having a methyl group instead
showed a 2-fold loss in activity. Since Tyr585 is exposed to solvent
and the side chain forms hydrogen bonds with water molecules,
the interaction between trifluoromethyl and Tyr585 is geometri-
cally imperfect. Thus the interaction seems to work weaker than
that in ideal hydrogen bonds. The distance from the carbon atom
of the guanidine group of Arg358 to the quinolyl ring is 3.3 Å,
9d
9e
9f
1.2
1.5
NC
NC
N
O
0.50
0.55
0.42
0.72
0.77
0.69
N
S
N
S
9g
Kim et al., reported that the guanidine moiety of Arg358 interacts
with the trifluoromethyl substituent of sitagliptin (1), which seems
to increase the activity and that much attention has focused on the
interaction with Arg358 in DPP-4 inhibitors recently.18 This result
suggests that the interaction with the S2 extensive subsite plays an
important role for DPP-4 inhibition.
which is slightly shorter than that of common CH-
This suggests that CH- interactions formed by induced fit be-
p
interactions.17
p
In vitro human and rat DPP-4 inhibitory activity of fused bicy-
clic heteroaryl, such as benzimidazole, benzoxazole and benzothi-
azole, are listed in Table 2. The activities of compounds 9a, 9d, and
tween the quinolyl ring and the guanidinyl group of Arg358 in-
creased the activity. We define the space consisting of Arg358
and surrounding amino acids as ‘S2 extensive subsite’ ( Fig. 3).
Figure 3. Binding interactions in active site. Co-crystallization of compound 8g and human DPP-4. The surface in pink represents S2 extensive subsite.

5036
T. Yoshida et al. / Bioorg. Med. Chem. 20 (2012) 5033–5041
Figure 4. Docking model of compound 9a and DPP-4. Docking model of compound 9a (magenta carbon) and DPP-4 is superimposed on the X-ray structure of 8g (cyan
carbon). Phe357 and Arg358 are analyzed with compound 8g. Arg358 (orange stick) is included in the X-ray structure ‘1NU8’19 in Protein Data Bank.
9f were similar to those of 8a and 8e. Figure 4 shows a docking
model of compound 9a and DPP-4 which is superimposed on the
X-ray structure of 8g. In this model, compound 9a lost the stacking
interaction with Phe357. Meanwhile, benzimidazole of 9a formed a
weak hydrogen bond with the side chain of Arg358 known for mul-
tiple conformations by many DPP-4 X-ray structures. Since the lost
interaction with Phe357 would be compensated by the flexibility
of Arg358, compounds 9a, 9d, and 9f seem to keep similar activi-
ties to 8a and 8e. Nitrile compounds 9c and 9e showed 2-fold high-
er activities than unsubstituted compounds 9a and 9d,
respectively. The docking model of 9e is shown in Figure 5. The ni-
trile group on the benzoxazole formed a direct hydrogen bond with
the side chain of Arg356. It might be expected to work much more
effectively, however, the penalty of the desolvation energy of
Arg356 making the hydrogen bond with the nitrile group has to
be taken into consideration. Moreover, Arg356 is observed to form
hydrogen bonds with the side chain of Glu403 and the main chain
of Val404 in the X-ray structure of DPP-4 complexed with com-
pound 8g, thus the conformation of Arg356 is rigid. Since the nitrile
group of compound 9g made no hydrogen bond with the side chain
of Arg356 in the docking model (data not shown), 9g is thought to
have comparable activity to 9f.
The enzymes most closely related to DPP-4 are the fibroblast acti-
vation protein (FAP), DPP-II, DPP-8 and DPP-9.13 Although the pre-
cise physiological functions of the enzymes are not known, DPP-8
and DPP-9 are widely distributed cytosolic enzymes, and their inhi-
bition would induce the toxicity of DPP-4 inhibitors identified to
date, including alopecia, thrombocytopenia, anaemia, enlarged
spleen, multiple histological pathologies, and animal mortality.14
The (4-nitrophenyl)piperazine compound 5b has moderate selectiv-
ity against DPP-8 and DPP-9 (8- and 19-fold, respectively), while the
quinolyl compound 8g showed at least 200-fold selectivity over
them (Table 3). Kang et al., reported that Phe357 plays an important
role in increasing the selectivity against DPP-9.20 Since both 5b and
8g form interactions with Phe357, there would be additional inter-
actions between 8g and DPP-4 to explain the high selectivity of 8g.
Amino acid sequences in the S1 subsite (Y547, S630, Y631, V656,
Y662, Y666, N710, V711 and H740 of DPP-4) are common between
DPP-4, DPP-8 and DPP-9,21 but those of the S2 extensive subsite
are different (Table 4), suggesting that the interaction formed by in-
duced fit between the quinolyl ring with the S2 extensive subsite
would contribute to the high selectivity of 8g against DPP-8 and
DPP-9. In order to increase the selectivity, the interaction with the
S2 extensive subsite should be carefully considered.
Figure 5. Docking model of compound 9e and DPP-4. Compound 9e (magenta carbon) forms a hydrogen bond with the side chain of Arg356 (orange carbon). Arg356 has
hydrogen bonds with Glu403 and Val404 (orange stick carbon).

T. Yoshida et al. / Bioorg. Med. Chem. 20 (2012) 5033–5041
5037
Table 3
doublet; t, triplet; q, quartet; m, multiplet; dd, double doublet;
dt, double triplet; br s, broad singlet. Electron analysis for carbon,
hydrogen, and nitrogen was performed with a Yanagimoto CHN
CORDER MT-6. LC–MS spectra were obtained on a PE-Sciex API
165 spectrometer with electrospray ionization (ESI) mode. Chro-
matography refers to flash chromatography conducted on silica
gel BW-300 (Fuji Silysia). All chemicals and solvents were of
reagent grade unless otherwise specified. For drying organic solu-
tions in extraction, anhydrous sodium sulfate or anhydrous mag-
nesium sulfate was used unless otherwise indicated. The
following abbreviations are used: DMF, N,N-dimethylformamide;
DMSO, dimethyl sulfoxide; EDC, 1-[3-(dimethylamino)propyl]-3-
ethylcarbodiimide hydrochloride; HOBT, 3-hydroxybenztriazole
hydrate.
Selectivity of DPP-4 inhibition against DPP-8 and DPP-9
Compound
Inhibitory activity IC50 (nmol/L) on human enzymes
DPP-4
DPP-8
DPP-9
5b
8g
1.6
12.2
72.4
30.2
105
0.37
Table 4
The alignment of the amino acids in S2 extensive subsite of DPP-4, DPP-8 and DPP-9
DPP-4 family
Corresponding amino acids in S2 extensive subsite
DPP-4
DPP-8
DPP-9
Arg358
Asp435
Asp425
Tyr585
His693
Gln684
5.1.2. 3-((S)-1-tert-Butoxycarbonyl-4-oxopyrrolidin-2-ylcar-
bonyl)thiazolidine (7a)
A mixture of N-tert-butoxycarbonyl-L-trans-hydroxyproline (6a)
0.3 µmol/kg,
1 µmol/kg
3 µmol/kg
(69.4 g, 0.3 mol), thiazolidine (29.4 g, 0.33 mmol), HOBT (50.5 g,
0.33 mol) and EDC (63.3 g, 0.33 mol) in DMF (300 mL) was stirred
at room temperature for 18 h. The reaction mixture was concen-
trated under reduced pressure. To the residue was added a satu-
rated aqueous sodium hydrogen carbonate solution and the
mixture was extracted with ethyl acetate. The extract was dried
and concentrated under reduced pressure to give 3-[(2S,4R)-1-
tert-butoxycarbonyl-4-hydroxypyrrolidin-2-ylcarbonyl]thiazoli-
dine (56.3 g, 62%) as a colorless oil. 1H NMR (300 MHz, CDCl3): d
1.41–1.45 (9H, m), 1.95–2.34 (2H, m), 2.62–3.25 (2H, m), 3.40–
3.98 (4H, m), 4.40–4.90 (4H, m).
100
80
60
40
20
To a solution of the above compound (55.4 g, 183 mmol) and
triethylamine (46 mL, 330 mmol) in dichloromethane (350 mL)
was added sulfur trioxide-pyridine complex (52.4 g, 329 mmol)
in DMSO (150 mL) under ice cooling and the mixture was stirred
for 2 h. The reaction mixture was poured into a saturated aqueous
sodium hydrogen carbonate solution and extracted with ethyl ace-
tate. The extract was washed with brine, dried and concentrated
under reduced pressure. The residue purified by silica gel chroma-
tography with n-hexane/ethyl acetate (1:1, v/v) to give the title
compound (30.3 g, 55%) as a white powder. 1H NMR (500 MHz,
DMSO-d6): d 1.36, 1.40 (9H, s), 2.36–2.45 (1H, m), 2.97–3.12 (3H,
m), 3.62–3.71 (2H, m), 3.74–3.94 (2H, m), 4.33–4.80 (2H, m),
4.91–5.04 (1H, m).
0
-2
0
2
4
6
8
10 12 14 16 18 20 22 24
Time after administaration (h)
Figure 6. Effect of 8g on plasma DPP-4 activity (% change of baseline) in Wistar rat.
8g was orally administrated at a dose of 0.3, 1 or 3
expressed as means SEM (n = 3).
lmol/kg at 0 h. Data are
Evaluation of plasma DPP-4 activity after oral dosing of the
compound 8g can be used to predict the efficacy of antihyperglyce-
mic activity and pharmacokinetic profile. The representative com-
pound 8g was administered orally to Wistar rats at a dose of 0.3, 1
or 3 lmol/kg and the plasma DPP-4 activity was evaluated ex vivo.
As shown in Figure 6, compound 8g Indicated dose-dependent,
fast-onset and long-lasting DPP-4 inhibitory activity. Most note-
worthy, more than 40% inhibition of plasma DPP-4 was sustained
5.1.3. 3-{(2S,4S)-4-[4-(Isoquinolin-1-yl)piperazin-1-yl]pyrroli-
din-2-ylcarbonyl}thiazolidine trihydrochloride (8a)
at a single oral dose of 3 lmol/kg until 24 h.
A mixture of 7a (606 mg, 2.02 mmol), 1-(isoquinolin-1-yl)piper-
azine (692 mg, 2.42 mmol) and sodium cyanoborohydride
(130 mg, 2.07 mmol) in methanol (10 mL) was stirred at room
temperature for 14 h. The reaction mixture was concentrated un-
der reduced pressure. The residue was poured into a saturated
aqueous sodium hydrogen carbonate solution and extracted with
chloroform. The extract was washed with brine, dried and concen-
trated under reduced pressure. The residue was purified by silica
gel chromatography with chloroform/methanol (20:1, v/v) to give
3-{(2S,4S)-1-tert-butoxycarbonyl-4-[4-(isoquinolin-1-yl)pipera-
zin-1-yl]pyrrolidin-2-ylcarbonyl}thiazolidine (216 mg, 21%) as a
pale-yellow powder.
4. Conclusion
A series of prolylthiazolidines with fused bicyclic heteroarylpip-
erazine substitution has been discovered as DPP-4 inhibitors, and
the most potent compound 8g showed high selectivity and
in vivo efficacy. X-ray crystal structure determination indicates
that CH-p interactions formed by induced fit between the quinolyl
ring and the guanidinyl group of Arg358 enhances the activity and
selectivity for DPP-4 inhibition.
5. Experimental
5.1. Chemistry
To a solution of the above compound (215 mg, 0432 mmol) in
methanol (3 mL) and chloroform (3 mL) was added 4 mol/L
hydrochloric acid in 1,4-dioxane (1 mL) and the mixture stirred
at room temperature overnight. The reaction mixture was con-
centrated under reduced pressure, and the residue was crystal-
lized with ethanol to give the title compound (99 mg, 42%) as a
pale-yellow powder. 1H NMR (500 MHz, DMSO-d6): d 2.25–2.30
(1H, m), 3.00–3.17 (3H, m), 3.59–3.95 (12H, m), 4.13–4.18 (1H,
m), 4.49–4.77 (3H, m), 7.59 (1H, d, J = 6.1 Hz), 7.71–7.74 (1H,
5.1.1. General
1H NMR spectra were measured on a Bruker DPX-300 instru-
ment or on a Bruker AMX-500 with tetramethylsilane as the inter-
nal standard; chemical shifts are reported in parts per million
(ppm, d units). Splitting patterns are designated as s, singlet; d,

5038
T. Yoshida et al. / Bioorg. Med. Chem. 20 (2012) 5033–5041
m), 7.86–7.89 (1H, m), 8.02 (1H, d, J = 8.2 Hz), 8.08 (1H, d,
J = 6.1 Hz), 8,21 (1H, d, J = 8.5 Hz), 9.25 (1H, br s), 10.89 (1H, br
s); Anal. Calcd for C21H27N5OSꢀ3HClꢀ0.3C2H6Oꢀ1.5H2O: C, 47.36;
H, 6.40; N, 12.79. Found: C, 47.26; H, 6.42; N, 12.60; LC–MS
(ESI) m/z 398.2 [M+H]+.
5.1.7. 3-{(2S,4S)-4-[4-(Quinolin-4-yl)piperazin-1-yl]pyrrolidin-
2-ylcarbonyl}thiazolidine trihydrochloride (8e)
The title compound was prepared in 34% yield using 1-(quino-
lin-4-yl)piperazine in the procedures outlined for 8b. 1H NMR
(300 MHz, DMSO-d6): d 2.16–2.40 (1H, m), 2.70–4.30 (16H, m),
4.40–4.80 (3H, m), 7.37 (1H, d, J = 6.9 Hz), 7.77 (1H, t, J = 8.1 Hz),
8.04 (1H, t, J = 8.4 Hz), 8.21 (2H, d, J = 8.7 Hz), 8.85 (1H, d,
J = 6.9 Hz); Anal. Calcd for C21H27N5OSꢀ3HClꢀ0.5C2H6Oꢀ2.2H2O: C,
46.39; H, 6.62; N, 12.30. Found: C, 46.34; H, 6.54; N, 12.47; LC–
MS (ESI) m/z 398.2 [M+H]+.
5.1.4. 3-{(2S,4S)-4-[4-(4-Chloroisoquinolin-1-yl)piperazin-1-yl]-
pyrrolidin-2-ylcarbonyl}thiazolidine hemipentahydrochloride
(8b)
To a solution of 7a (450 mg, 1.50 mmol), 1-(4-chloroisoquino-
lin-1-yl)piperazine (446 mg, 1.80 mmol) and acetic acid
(0.090 mL, 1.6 mmol) in 1,2-dichloroethane (8 mL) was added so-
dium triacetoxyborohydride (636 mg, 3.00 mmol) and the mixture
was stirred at room temperature for 3 h. The reaction mixture was
poured into a saturated aqueous sodium hydrogen carbonate solu-
tion and extracted with chloroform. The extract was washed with
brine, dried and concentrated under reduced pressure. The residue
purified by silica gel chromatography with chloroform/methanol
(50:1, v/v) to give 3-{(2S,4S)-1-tert-butoxycarbonyl-4-[4-(4-chlo-
roisoquinolin-1-yl)piperazin-1-yl]pyrrolidin-2-ylcarbonyl}thiazol-
idine (596 mg, 75%) as a white powder.
5.1.8. 3-{(2S,4S)-4-[4-(2-Methylquinolin-4-yl)piperazin-1-yl]-
pyrrolidin-2-ylcarbonyl}thiazolidine trihydrochloride (8f)
The title compound was prepared in 75% yield using 1-(2-meth-
ylquinol-4-yl)piperazine in the procedures outlined for 8b. 1H NMR
(300 MHz, DMSO-d6): d 2.20–2.42 (1H, m), 2.81 (3H, s), 2.91–3.20
(3H, m), 3.30–4.26(13H, m), 4.44–4.87 (3H, m), 7.35 (1H, s), 7.73
(1H, t, J = 7.6 Hz), 8.00 (1H, t, J = 7.6 Hz), 8.16 (1H, d, J = 8.4 Hz),
8.24 (1H, d, J = 8.4 Hz); Anal. Calcd for C22H29N5OSꢀ3HClꢀ0.6-
C2H6Oꢀ1.4H2O: C, 48.56; H, 6.75; N, 12.21. Found: C, 48.64; H,
6.72; N, 12.11; LC–MS (ESI) m/z 412.4 [M+H]+.
The above compound (592 mg, 1.11 mmol) was dissolved in
1.1 mol/L hydrogen chloride in methanol (10 mL), and the mixture
was stirred at room temperature for 5 days. The reaction mixture
was concentrated under reduced pressure, and the residue was
crystallized with ethanol to give the title compound (318 mg,
52%) as a pale-yellow powder. 1H NMR (300 MHz, DMSO-d6): d
2.32–2.46 (1H, m), 2.95–4.20 (16H, m), 4.43–4.78 (3H, m), 7.74–
7.82 (1H, m), 7.90–7.97 (1H, m), 8.14 (1H, d, J = 8.0 Hz), 8.23 (1H,
d, J = 8.3 Hz), 8.30 (1H, s), 9.17 (1H, br s), 10.83 (1H, br s), 12.53
(1H, br s); Anal. Calcd for C21H26ClN5OSꢀ2.5HClꢀ1.5H2O: C, 45.85;
H, 5.77; N, 12.73. Found: C, 45.59; H, 6.05; N, 12.82; LC–MS (ESI)
m/z 432.2 [M+H]+.
5.1.9. 3-{(2S,4S)-4-[4-(2-Trifluoromethylquinolin-4-yl)pipera-
zin-1-yl]pyrrolidin-2-ylcarbonyl}thiazolidine dihydrochloride
(8g)
The title compound was prepared in 47% yield using 1-(2-triflu-
oromethylquinolin-4-yl)piperazine in the procedures outlined for
8b. 1H NMR (500 MHz, DMSO-d6): d 2.41–2.43 (1H, m), 3.08–3.18
(3H, m), 3.57–3.98 (12H, m), 4.20–4.24 (1H, m), 4.50–4.81 (3H,
m), 7.39 (1H, s), 7.76 (1H, t, J = 7.6 Hz), 7.88–7.91 (1H, m), 8.12–
8.17 (2H, m), 9.26 (1H, br s), 11.10 (1H, br s); Anal. Calcd for
C
22H26F3N5OSꢀ2HClꢀ3H2O: C, 44.60; H, 5.78; N, 11.82. Found: C,
44.85; H, 5.73; N, 11.87; LC–MS (ESI) m/z 466.4 [M+H]+.
5.1.5. 3-{(2S,4S)-4-[4-(4-Cyanoisoquinolin-1-yl)piperazin-1-yl]-
pyrrolidin-2-ylcarbonyl}thiazolidine Trihydrochloride (8c)
1-Chloro-4-cyanoisoquinoline (500 mg, 2.65 mmol) was added
to piperazine (4.61 g, 53.5 mmol) at 140 °C. The mixture was stir-
red at 140 °C for 2 h, and then water was added to the reaction
mixture. The mixture was extracted with chloroform, and the ex-
tract was washed with brine, dried and concentrated under re-
duced pressure to give 1-(4-cyanoisoquinolin-1-yl)piperazine
(491 mg, 78%) as a dark brown solid. 1H NMR (300 MHz, CDCl3):
d 3.08–3.17 (4H, m), 3.62–3.68 (4H, m), 7.59 (1H, t, J = 7.1 Hz),
7.77 (1H, t, J = 7.1 Hz), 8.02–8.09 (2H, m), 8.47 (1H, s).
5.1.10. 3-{(2S,4S)-4-[4-(7-Chloroquinolin-4-yl)piperazin-1-yl]-
pyrrolidin-2-ylcarbonyl}thiazolidine trihydrochloride (8h)
The title compound was prepared in 55% yield using 1-(7-chlo-
roquinolin-4-yl)piperazine in the procedures outlined for 8b. 1H
NMR (300 MHz, DMSO-d6): d 2.10–2.37 (1H, m), 2.84–4.00 (16H,
m), 4.41–4.82 (3H, m), 7.36 (1H, d, J = 6.9 Hz), 7.77 (1H, dd,
J = 1.8, 9.0 Hz), 8.22 (1H, d, J = 9.0 Hz), 8.27 (1H, d, J = 1.8 Hz),
8.85 (1H, d, J = 6.9 Hz), 9.18 (1H, br s), 10.82 (1H, br s); Anal. Calcd
for C21H26ClN5OSꢀ3HClꢀ1.5H2O: C, 44.38; H, 5.67; N, 12.32. Found:
C, 44.24; H, 5.71; N, 12.23; LC–MS (ESI) m/z 432.2 [M+H]+.
The title compound was prepared in 36% yield using the above
compound in the procedures outlined for 8b. 1H NMR (300 MHz,
DMSO-d6): d 2.26–2.43 (1H, m), 2.93–4.20 (16H, m), 4.44–4.78
(3H, m), 7.74–7.82 (1H, m), 7.75–8.05 (2H, m), 8.22 (1H, d,
J = 8.4 Hz), 8.69 (1H, s), 9.16 (1H, br s), 10.85 (1H, br s), 12.65
(1H, br s); Anal. Calcd for C22H26N6OSꢀ3HClꢀ0.4C2H6OꢀH2O: C,
48.18; H, 5.92; N, 14.79. Found: C, 48.02; H, 6.16; N, 15.13; LC–
MS (ESI) m/z 423.2 [M+H]+.
5.1.11. 3-{(2S,4S)-4-[4-(7-Trifluoromethylquinolin-4-yl)pipera-
zin-1-yl]pyrrolidin-2-ylcarbonyl}thiazolidine Trihydrochloride
(8i)
The title compound was prepared in 23% yield using 1-(7-triflu-
oromethylquinolin-4-yl)piperazine in the procedures outlined for
8b. 1H NMR (300 MHz, DMSO-d6): d 2.20–2.47 (1H, m), 2.90–3.20
(3H, m), 3.30–4.30 (13H, m), 4.45–4.85 (3H, m), 7.46 (1H, d,
J = 6.7 Hz), 7.98 (1H, dd, J = 1.5, 9.0 Hz), 8.43 (1H, d, J = 8.9 Hz),
8.62 (1H, s), 8.96 (1H, d, J = 6.7 Hz); Anal. Calcd for
5.1.6. 3-{(2S,4S)-4-[4-(Quinolin-2-yl)piperazin-1-yl]pyrrolidin-
2-ylcarbonyl}thiazolidine trihydrochloride (8d)
C
22H26F3N5OSꢀ3HClꢀ0.5C2H6OꢀH2O: C, 44.85; H, 5.56; N, 11.37.
Found: C, 44.94; H, 5.51; N, 11.38; LC–MS (ESI) m/z 466.4 [M+H]+.
The title compound was prepared in 61% yield using 1-(quino-
lin-2-yl)piperazine in the procedures outlined for 8b. 1H NMR
(300 MHz, DMSO-d6): d 2.20–2.30 (1H, m), 2.96–3.17 (3H, m),
3.64–4.40 (13H, m), 4.47–4.76 (3H, m), 7.50 (1H, t, J = 7.5 Hz),
7.60 (1H, d, J = 9.6 Hz), 7.77 (1H, t, J = 7.8 Hz), 7.93 (1H, d,
J = 7.5 Hz), 8.15–8.20 (1H, m), 8.44 (1H, d, J = 9.6 Hz), 9.21 (1H, br
s), 10.68 (1H, br s); Anal. Calcd for C21H27N5OSꢀ3HClꢀ3.5H2O: C,
44.25; H, 6.54; N, 12.29. Found: C, 44.23; H, 6.34; N, 12.18; LC–
MS (ESI) m/z 398.2 [M+H]+.
5.1.12. 3-{(2S,4S)-4-[4-(Benzimidazol-2-yl)piperazin-1-yl]pyrr-
olidin-2-ylcarbonyl}thiazolidine dihydrochloride (9a)
The title compound was prepared in 43% yield using 1-(ben-
zimidazol-2-yl)piperazine in the procedures outlined for 8b. 1H
NMR (300 MHz, DMSO-d6): d 1.72–2.16 (1H, m), 2.65–4.30 (16H,
m), 4.40–4.80 (3H, m), 7.18–7.33 (2H, m), 7.36–7.51 (2H, m),
8.95 (1H, br s), 9.70 (1H, br s), 10.50 (1H, br s), 13.71 (1H, br s);
Anal. Calcd for C19H26N6OSꢀ2HClꢀ3H2O: C, 44.44; H, 6.67; N,

T. Yoshida et al. / Bioorg. Med. Chem. 20 (2012) 5033–5041
5039
16.37. Found: C, 44.67; H, 6.52; N, 16.73; LC–MS (ESI) m/z 387.2
a purple solid. 1H NMR (500 MHz, DMSO-d6): d 7.07 (1H, d,
J = 8.2 Hz), 7.31 (1H, s), 7.39 (1H, d, J = 8.2 Hz), 11.03 (1H, br s),
11.16 (1H, br s); LC–MS (ESI) m/z 160.0 [M+H]+.
[M+H]+.
5.1.13. 3-[(2S,4S)-1-Benzyloxycarbonyl-4-(piperazin-1-yl)pyrr-
olidin-2-ylcarbonyl]thiazolidine (10)
A mixture of the above compound (894 mg, 5.62 mmol) and
phosphorus oxychloride (12 mL) was refluxed for 3 h. The reaction
mixture was added to ice, and extracted with ethyl acetate. The
extract was washed with brine, dried and concentrated under re-
duced pressure. The residue was purified by silica gel chromatog-
raphy with n-hexane/ethyl acetate (2:3, v/v) to give the title
compound (322 mg, 32%) as a white powder. LC–MS (ESI) m/z
178.0 [M+H]+.
To a solution of 3-((S)-1-benzyloxycarbonyl-4-oxopyrrolidin-2-
ylcarbonyl)thiazolidine (7b) (1.36 g, 4.07 mmol), 1-(tert-butoxy-
carbonyl)piperazine (910 mg, 4.89 mmol) and acetic acid
(0.24 mL, 4.2 mmol) in 1,2-dichloroethane (40 mL) was added so-
dium triacetoxyborohydride (1.72 g, 8.11 mmol) and the mixture
was stirred at room temperature overnight. The reaction mixture
was poured into a saturated aqueous sodium hydrogen carbonate
solution and extracted with chloroform. The extract was washed
with brine, dried and concentrated under reduced pressure. The
residue was purified by silica gel chromatography with ethyl
acetate to give 3-[(2S,4S)-1-benzyloxycarbonyl-4-(4-tert-butoxy-
carbonylpiperazin-1-yl)pyrrolidin-2-ylcarbonyl]thiazolidine (1.80 g,
88%) as a white powder.
A mixture of the above compound (1.79 g, 3.55 mmol) and tri-
fluoroacetic acid (10 mL) in dichloromethane (100 mL) was stirred
at room temperature overnight. The reaction mixture was concen-
trated under reduced pressure, and the residue was poured into a
saturated aqueous sodium hydrogen carbonate solution and ex-
tracted with chloroform. The extract was washed with brine, dried
and concentrated under reduced pressure to give the title com-
pound (1.38 g, 96%) as a white powder. 1H NMR (300 MHz,
DMSO-d6): d 1.46–1.55 (1H, m), 1.90–3.14 (13H, m), 3.52–3.88
(3H, m), 4.33–4.72 (3H, m), 4.87–5.08 (2H, m), 7.27–7.43 (5H,
m); LC–MS (ESI) m/z 405.4 [M+H]+.
5.1.16.3-{(2S,4S)-4-[4-(5-Cyanobenzimidazol-2-yl)piperazin-1-yl]
pyrrolidin-2-ylcarbonyl}thiazolidine Trihydrochloride (9c)
3-{(2S, 4S)-1-Benzyloxycarbonyl-4-[4-(5-cyanobenzimidazol-2-yl)
piperazin-1-yl]pyrrolidin-2-ylcarbonyl}thiazolidine was prepared in
54% yield using 11b in the procedures outlined for 9b.
A mixture of the above compound (200 mg, 0.367 mmol) and
thioanisole (0.43 mL, 3.7 mmol) in trifluoroacetic acid (10 mL)
was stirred at room temperature overnight. Diethyl ether was
added to the reaction mixture, and the precipitate was collected
by filtration and purified by high performance liquid chromatogra-
phy and converted to hydrochloride salt with 4 mol/L hydrogen
chloride in ethyl acetate to give the title compound (50 mg, 26%)
as a white powder. 1H NMR (500 MHz, DMSO-d6): d 1.89–2.09
(1H, m), 2.78–4.20 (16H, m), 4.47–4.82 (3H, m), 7.51 (1H, d,
J = 8.2 Hz), 7.61 (1H, d, J = 8.2 Hz), 7.81 (1H, s), 8.97 (1H, br s),
10.28 (1H, br s); Anal. Calcd for
C
20H25N7OSꢀ3HClꢀ0.4-
C4H8O2ꢀ2.7H2O: C, 42.90; H, 6.10; N, 16.21. Found: C, 42.92; H,
5.82; N, 16.12; LC–MS (ESI) m/z 412.4 [M+H]+.
5.1.14. 3-{(2S,4S)-4-[4-(5-Chlorobenzimidazol-2-yl)piperazin-1-
yl]pyrrolidin-2-ylcarbonyl}thiazolidine Trihydrobromide (9b)
A mixture of 10 (345 mg, 0.853 mmol), 2,5-dichlorobenzimi-
dazole (11a) (191 mg, 1.02 mmol) and N,N-diisopropylethylamine
(0.18 mL, 1.0 mmol) in N-methyl-2-pyrrolidone (6 mL) was stirred
at 100 °C overnight. The reaction mixture was poured into water
and extracted with ethyl acetate. The extract was washed with
brine, dried and concentrated under reduced pressure. The residue
was purified by high performance liquid chromatography to give
3-{(2S,4S)-1-benzyloxycarbonyl-4-[4-(5-chlorobenzimidazol-2-
5.1.17. 3-{(2S,4S)-4-[4-(Benzoxazol-2-yl)piperazin-1-yl]pyrr-
olidin-2-ylcarbonyl}thiazolidine trihydrochloride (9d)
The title compound was prepared in 34% yield using 1-(ben-
zoxazol-2-yl)piperazine in the procedures outlined for 8b. 1H
NMR (300 MHz, DMSO-d6): d 2.20–2.42 (1H, m), 2.89–3.20 (3H,
m), 3.25–4.35 (13H, m), 4.40–4.80 (3H, m), 7.10 (1H, dt, J = 1.2,
7.5 Hz), 7.22 (1H, dt, J = 1.2, 7.8 Hz), 7.37 (1H, dd, J = 0.6,
7.8 Hz), 7.47 (1H, d, J = 7.8 Hz), 9.25 (1H, br s), 11.00 (1H, br s);
Anal. Calcd for C19H25N5O2Sꢀ3HClꢀ0.3C2H6Oꢀ1.5H2O: C, 43.78; H,
6.15; N, 13.02. Found: C, 43.74; H, 6.23; N, 13.18; LC–MS (ESI)
m/z 388.4 [M+H]+.
yl)piperazin-1-yl]pyrrolidin-2-ylcarbonyl}thiazolidine
26%) as a pale-yellow powder.
(122 mg,
A mixture of the above compound (110 mg, 0.198 mmol) and
30% hydrogen bromide-acetic acid solution (10 mL) was stirred
at room temperature for 6 h. Diethyl ether was added to the
reaction mixture, and the precipitate was collected by filtration
and recrystallized with ethanol to give the title compound
(56 mg, 40%) as a white powder. 1H NMR (500 MHz, DMSO-d6):
5.1.18. 3-{(2S,4S)-4-[4-(5-Cyanobenzoxazol-2-yl)piperazin-1-
yl]pyrrolidin-2-ylcarbonyl}thiazolidine Dihydrochloride (9e)
The title compound was prepared in 28% yield using 1-(5-cya-
nobenzoxazol-2-yl)piperazine in the procedures outlined for 8b.
1H NMR (300 MHz, DMSO-d6): d 1.94–2.26 (1H, m), 2.80–3.00
(1H, m), 3.00–4.30 (15H, m), 4.45–4.78 (3H, m), 7.56 (1H, dd,
J = 1.8, 8.4 Hz), 7.66 (1H, d, J = 8.4 Hz), 7.83 (1H, d, J = 1.5 Hz),
9.05 (1H, br s), 10.43 (1H, br s); Anal. Calcd for
d
1.76–1.96 (1H, m), 2.70–4.87 (16H, m), 4.46–4.74 (3H, m),
7.32 (1H, dd, J = 1.7, 8.5 Hz), 7.44 (1H, d, J = 8.5 Hz), 7.48 (1H,
d, J = 1.7 Hz), 8.96 (1H, br s), 9.59 (1H, br s), 13.15 (1H, br s);
Anal. Calcd for C19H25ClN6OSꢀ3HBrꢀ2H2O: C, 32.61; H, 4.61; N,
12.01. Found: C, 32.65; H, 4.38; N, 11.82; LC–MS (ESI) m/z
421.2 [M+H]+.
C
20H24N6O2Sꢀ2HClꢀ0.5H2O: C, 48.58; H, 5.50; N, 17.00. Found: C,
48.33; H, 5.58; N, 16.71; LC–MS (ESI) m/z 413.4 [M+H]+.
5.1.19. 3-{(2S,4S)-4-[4-(Benzothiazol-2-yl)piperazin-1-yl]pyrr-
olidin-2-ylcarbonyl}thiazolidine trihydrochloride (9f)
5.1.15. 2-Chlorobenzimidazole-5-carbonitrile (11b)
A solution of 3,4-diaminobenzonitrile (2.60 g, 18.2 mmol) and
pyridine (2 mL, 25 mmol) in DMF (20 mL) was added a solution
of bis(trichloromethyl)carbonate (2.12 g, 7.14 mmol) in tetrahy-
drofuran (20 mL) dropwise under ice-cooling. The mixture was
stirred at room temperature for 18 h. Dilute hydrochloric acid
was added to the reaction mixture, and the mixture was extracted
with ethyl acetate. The extract was washed with brine, dried and
concentrated under reduced pressure. To the residue was added
ethyl acetate, and then the precipitate was collected by filtration
to give 2-hydroxybenzimidazole-5-carbonitrile (896 mg, 29%) as
The title compound was prepared in 64% yield using 1-(ben-
zothiazol-2-yl)piperazine in the procedures outlined for 8b. 1H
NMR (300 MHz, DMSO-d6): d 2.20–2.44 (1H, m), 2.90–3.20 (3H,
m), 3.35–4.30 (13H, m), 4.42–4.82 (3H, m), 7.16 (1H, t,
J = 7.2 Hz), 7.35 (1H, t, J = 7.2 Hz), 7.55 (1H, d, J = 7.8 Hz), 7.86
(1H, d, J = 7.5 Hz), 9.25 (1H, br s), 10.90 (1H, br s); Anal. Calcd
for
C
19H25N5OS2ꢀ3HClꢀ0.2C2H6OꢀH2O: C, 43.14; H, 5.82; N,
12.96. Found: C, 43.07; H, 5.70; N, 12.99; LC–MS (ESI) m/z
404.4 [M+H]+.

5040
T. Yoshida et al. / Bioorg. Med. Chem. 20 (2012) 5033–5041
5.1.20. 3-{(2S,4S)-4-[4-(5-Cyanobenzothiazol-2-yl)piperazin-1-
yl]pyrrolidin-2-ylcarbonyl}thiazolidine Dihydrochloride (9g)
The title compound was prepared in 48% yield using 1-(5-cya-
nobenzothiazol-2-yl)piperazine in the procedures outlined for 8b.
1H NMR (300 MHz, DMSO-d6): d 2.04–2.28 (1H, m), 2.82–3.00
(1H, m), 3.00–4.30 (15H, m), 4.43–4.80 (3H, m), 7.53 (1H, dd,
J = 1.5, 8.2 Hz), 7.93 (1H, d, J = 1.5 Hz), 8.07 (1H, d, J = 8.2 Hz),
9.08 (1H, br s), 10.51 (1H, br s); Anal. Calcd for
5.4. Biological experiments
5.4.1. DPP-4 inhibitory activity
The DPP-4 inhibitory activity of human and rat plasma was
measured by fluorescence assay using Gly-Pro-MCA (Peptide Insti-
tute Inc.) as a DPP-4-specific fluorescent substrate. Reaction solu-
tions containing 20
solution), 20 L of fluorescent substrate (100
buffer (0.003% Brij-35 containing PBS), and 20 l
l
L of human or rat plasma (10-fold diluted
mol/L), 140 L of
L of test substrate
l
l
l
C
20H24N6OS2ꢀ2HClꢀH2O: C, 46.24; H, 5.43; N, 16.18. Found: C,
46.05; H, 5.41; N, 16.04; LC–MS (ESI) m/z 429.2 [M+H]+.
(of various concentrations) were incubated at room temperature
for 60 min using a 96-well flat-bottomed microtiter plate. The
measured fluorescent intensity (excitation 360 nm/emission
465 nm, SPECTRA FLUOR, TECAN) was taken as the DPP-4 activity.
The inhibitory rate relative to the solvent addition group was cal-
culated and IC50 values were determined by logistic analysis.
5.2. X-Ray crystallographic study
The protein of human DPP-4 (33ꢁ766) secreted from insect cells
was purified and crystallized according to the method reported by
Hiramatsu et al.22 The protein-inhibitor complex was obtained by
soaking a preformed DPP-4 crystal in the presence of compound
8g and preserved in liquid nitrogen for data collection at 100 K.
X-ray diffraction data were collected at the high energy accelerator
research organization (KEK) beam line BL5 and processed using the
program HKL2000.23 The structure of DPP-4 inhibitor complex was
solved by molecular replacement with the program PHASER,24 uti-
lizing the previously determined coordinates of DPP-4 with Protein
Data Bank accession code 1J2E. The structure was refined against
all available data to 2.10 Å using Maximum likelihood (Refmac)
to a crystallographic R-factor of 18.5% and free R-factor of 22.4%.
Coordinates have been deposited with the Protein Data Bank,
accession code 3VJM. Data collection and model refinement
statistics are summarized in Table 5.
5.4.2. DPP-8,9 inhibitory activity
DPP-8 and DPP-9 enzymes were prepared from cytoplasmic
fractions of cells expressing recombinant human DPP-8 or DPP-9,
respectively. Reaction solutions containing 20
pounds of various concentrations, 20 L of enzyme preparations,
140 L of buffer (0.003% Brij-35 containing PBS), and 20 L of
Gly-Pro-MCA (50 mol/L, Peptide Institute Inc.) were incubated
lL of test com-
l
l
l
l
at 37 °C for 30 min using a 96-well flat-bottomed microtiter plate.
The measured fluorescence intensity of 7-amino-4-methylcouma-
rin was taken as the enzyme.
5.4.3. Plasma DPP-4 activity after oral administration of 8g to
Wistar rats
Male Wistar rats (7–9 weeks of age) fasted overnight were used.
Compound 8g was dissolved in 0.5% hydroxypropylmethyl-cellu-
5.3. Docking studies in DPP-4
lose and administered orally at a dose of 0.3, 1 or 3 lmol/kg. At
pre-administration and at 0.5, 1, 2, 3, 5, 7, 9, and 24 h after admin-
istration, 0.1 mL of blood was collected from jugular vein. After
The x-ray crystal structure of DPP-4 complexed with compound
8g was utilized in the docking calculations. The compounds were
docked into DPP-4 using Glide 5.7.25
centrifugation, 10
(0.003% Brij-35 containing PBS). 20
used instead of 20 L of test substrate for the determination of
lL of plasma was diluted 10-fold using buffer
l
L of the diluted plasma was
Table 5
l
DPP-4 complex with 8g data collection and refinement statistics
DPP-4 inhibitory activity by fluorescence as described above.
PDB entry code
3VJM
Crystal
Acknowledgments
Space group
Unit cell parameters: a (Å)
b (Å)
c (Å)
Data
P212121
117.95
125.94
137.21
We thank Dr. Kazutoshi Watanabe and Dr. Kunitomo Adachi for
helpful discussions. We also thank Dr. Tohru Nakajima and Dr. Tak-
ao Kondo for their insight and guidance for course of this work.
Resolution (Å)
Unique reflections
Redundancy
Completeness (%)
50.00–2.10 (2.18–2.10)
119422 (10763)
4.8 (4.5)
91.0 (91.2)
0.067 (0.399)
11.7
References and notes
a
Rmerge
1. World Health Organization. 2011 <http://www.who.int/diabetes/en/>.
2. Bolen, S.; Feldman, L.; Vassy, J.; Wilson, L.; Yeh, H. C.; Marinpoulos, S.; Wiley, C.;
Selvin, E.; Wilson, R.; Bass, E. B.; Brancati, F. L. Ann. Intern. Med. 2007, 147, 386.
3. (a) Kieffer, T. J.; McIntosh, C. H.; Pederson, R. A. Endocrinology 1995, 136, 3585;
(b) Deacon, C. F.; Nauck, M. A.; Toft-Nielson, M.; Pridal, L.; Willms, B.; Holst, J. J.
Diabetes 1995, 44, 1126.
4. Pospisilik, J. A.; Stafford, S. G.; Demuth, H.-U.; Brownsey, R.; Arkhouse, W.;
Finegood, D. T.; McIntosh, C. H. S.; Pederson, R. A. Diabetes 2002, 51, 943.
5. (a) Idris, I.; Donnelly, R. Diabetes Obes. Metab. 2007, 9, 153; (b) von Geldern, T.
W.; Trevillyan, J. M. Drug Dev. Res. 2006, 67, 627; (c) Weber, A. E. J. Med. Chem.
2004, 47, 4135.
I/
r
Refinement
Resolution (Å)
Unique reflections
Completeness (%)
Data in the test set
R-work
30.00–2.10 (2.15–2.10)
109644 (7466)
91.8 (90.6)
5446 (411)
0.187 (0.225)
0.224 (0.277)
R-free
Structure
Protein non-H atoms/B (Å2)
Ligand atoms/B (Å2)
Water oxygen atoms/B (Å2)
Rmsd
12180/28.4
64/27.6
987/31.3
6. Gallwitz, B. Expert Opin. Invest. Drugs 2011, 20, 723.
7. Havale, S. H.; Pal, M. Bioorg. Med. Chem. 2009, 17, 1783.
8. (a) Kelly, T. A.; Adams, J.; Bachovchin, W. W.; Barton, R. W.; Campbell, S. J.;
Ciutts, S. J.; Kennedy, C. A.; Snow, R. J. J. Am. Chem. Soc. 1993, 115, 12637; (b)
Peters, J.-U. Curr. Top. Med. Chem. 2007, 7, 579.
9. Ashworth, D. M.; Atrash, B.; Baker, G. R.; Baxter, A. J.; Jenkins, P. D. Bioorg. Med.
Chem. Lett. 1996, 6, 1163.
10. (a) Magnin, D. R.; Robl, J. A.; Sulsky, R. B.; Augeri, D. J.; Huang, Y.; Simpkins, L.
M.; Taunk, P. C.; Betebenner, D. A.; Robertson, J. G.; Abboa-Offei, B. E.; Wang, A.;
Cap, M.; Xin, L.; Tao, L.; Sitkoff, D. F.; Malley, M. F.; Gougoutas, J. Z.; Khanna, A.;
Huang, Q.; Han, S.-P.; Parker, R. A.; Hamann, L. G. J. Med. Chem. 2004, 47, 2587;
(b) Augeri, D. J.; Robl, J. A.; Betebenner, D. A.; Magnin, D. R.; Khanna, A.;
Robertson, J. S. G.; Wang, A.; Simpkins, L. M.; Taunk, P.; Huang, Q.; Han, S.-P.;
Bond lengths (Å)
Bond angles (°)
Ramachandran plot
Most favored regions (%)
Additionally allowed regions (%)
Generously allowed regions (%)
0.011
1.306
89.2
10.2
0.6
Values in parentheses are for highest-resolution shell.
a
Rmerge
=
R
|(Iꢁ<I>)|/
R(I), where I is the observed intensity.

T. Yoshida et al. / Bioorg. Med. Chem. 20 (2012) 5033–5041
5041
Abboa-Offei, B.; Cap, M.; Xin, L.; Tao, L.; Tozzo, E.; Welzel, G. E.; Egan, D. M.;
Marcinkeviciene, J.; Chang, S. Y.; Biller, S. A.; Kirby, M. S.; Parker, R. A.; Hamann,
L. G. J. Med. Chem. 2005, 48, 5025.
16. Howard, J. A. K.; Hoy, V. J.; O’Hagan, D.; Smith, G. T. Tetrahedron 1996, 52,
12613.
17. Brandl, M.; Weiss, M. S.; Jabs, A.; Sühnel, J.; Hilgenfeld, R. J. Mol. Biol. 2001, 307,
357.
18. Kim, D.; Wang, L.; Beconi, M.; Eiermann, G. J.; Fisher, M. H.; He, H.; Hickey, G. J.;
Kowalchick, J. E.; Leiting, B.; Lyons, K.; Marsilio, F.; McCann, M. E.; Patel, R. A.;
Petrov, A.; Scapin, G.; Patel, S. B.; Roy, R. S.; Wu, J. K.; Wyvratt, M. J.; Zhang, B.
B.; Thornberry, N. A.; Weber, A. E. J. Med. Chem. 2005, 48, 141.
19. Thoma, R.; Löffler, B.; Stihle, M.; Huber, W.; Ruf, A.; Hennig, M. Structure 2003,
11, 947.
20. Kang, N. S.; Ahn, J. H.; Kim, S. S.; Chae, C. H.; Yoo, S.-E. Bioorg. Med. Chem. Lett.
2007, 17, 3716.
21. Rummey, C.; Metz, G. Proteins 2007, 66, 160.
22. Hiramatsu, H.; Kyono, K.; Shima, H.; Fukushima, C.; Sugiyama, S.; Inaka, K.;
Yamamoto, A.; Shimizu, R. Acta Crystallogr. 2003, D59, 595.
23. Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
24. McCoy, A. J.; Grosse-Kunstleve, R. W.; Storoni, L. C.; Read, R. J. Acta Crystallogr.
2005, D61, 458.
11. (a) Sakashita, H.; Akahoshi, F.; Kitajima, H.; Tsutsumiuchi, R.; Hayashi, Y.
Bioorg. Med. Chem. 2006, 14, 3662; (b) Sakashita, H.; Akahoshi, F.; Yoshida, T.;
Kitajima, H.; Hayashi, Y.; Ishii, S.; Takashina, Y.; Tsutsumiuchi, R.; Ono, S.
Bioorg. Med. Chem. 2007, 15, 641.
12. Yoshida, T.; Sakashita, H.; Akahoshi, F.; Hayashi, Y. Bioorg. Med. Chem. Lett.
2007, 17, 2618.
13. (a) Park, J. E.; Lenter, M. C.; Zimmermann, R. N.; Garin-Chesa, P.; Old, L. J.;
Rettig, W. J. J. Biol. Chem. 1999, 274, 36505; (b) Abbott, C. A.; Yu, D. M.;
Woollatt, E.; Sutherland, G. R.; McCaughan, G. W.; Gorrell, M. D. Eur. J. Biochem.
2000, 267, 6140; (c) Ajami, K.; Abbott, C. A.; McCaughan, G. W.; Gorrell, M. D.
Biochim. Biophys. Acta 2004, 1679, 18.
14. Lankas, G. R.; Leiting, B.; Roy, R. S.; Eiermann, G. J.; Beconi, M. G.; Biftu, T.; Chan,
C. C.; Edmondson, S.; Feeney, W. P.; He, H.; Ippolito, D. E.; Kim, D.; Lyons, K. A.;
Ok, H. O.; Patel, R. A.; Petrov, A. N.; Pryor, K. A.; Qian, X.; Reigle, L.; Woods, A.;
Wu, J. K.; Zaller, D.; Zhang, X.; Zhu, L.; Weber, A. E.; Thornberry, N. A. Diabetes
2005, 54, 2988.
25. Glide; Schrodinger, LLC, New York, NY. 2011 <http://www.schrodinger.com/>.
15. Parikh, R.; Doering, W. V. E. Arn. Soc. 1967, 89, 5505.