Design and synthesis of a series of α-benzyl phenylpropanoic acid-type peroxisome proliferator-activated receptor (PPAR) gamma partial agonists with improved aqueous solubility

Article abstract of DOI:10.1016/j.bmc.2013.02.003

In the continuing study directed toward the development of peroxisome proliferator-activated receptor gamma (hPPARγ) agonist, we attempted to improve the water solubility of our previously developed hPPARγ-selective agonist 3, which is insufficiently soluble for practical use, by employing two strategies: introducing substituents to reduce its molecular planarity and decreasing its hydrophobicity via replacement of the adamantyl group with a heteroaromatic ring. The first approach proved ineffective, but the second was productive. Here, we report the design and synthesis of a series of α-benzyl phenylpropanoic acid-type hPPARγ partial agonists with improved aqueous solubility. Among them, we selected (R)-7j, which activates hPPARγ to the extent of about 65% of the maximum observed with a full agonist, for further evaluation. The ligand-binding mode and the reason for the partial-agonistic activity are discussed based on X-ray-determined structure of the complex of hPPARγ ligand-binding domain (LBD) and (R)-7j with previously reported ligand-LDB structures. Preliminal apoptotic effect of (R)-7j against human scirrhous gastric cancer cell line OCUM-2MD3 is also described.

Full text of DOI:10.1016/j.bmc.2013.02.003

Bioorganic & Medicinal Chemistry 21 (2013) 2319–2332
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design and synthesis of a series of a-benzyl phenylpropanoic
acid-type peroxisome proliferator-activated receptor (PPAR)
gamma partial agonists with improved aqueous solubility
Masao Ohashi a, Takuji Oyama b, Endy Widya Putranto c, Tsuyoshi Waku d, Hiromi Nobusada a,
Ken Kataoka e, Kenji Matsuno a, Masakazu Yashiro f, Kosuke Morikawa g, , Nam-ho Huh c,
Hiroyuki Miyachi a,
⇑
a Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, 1-1-1 Tsushima-Naka, Kita-ku, Okayama 700-8530, Japan
b Department of Biotechnology, Faculty of Life and Environmental Sciences, University of Yamanashi, 4-4-37 Takeda, Kofu City, Yamanashi 400-8510, Japan
c Department of Cell Biology, Okayama University, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 2-5-1 Shikatacho, Kita-ku, Okayama 700-8558, Japan
d Graduate School of Pharmaceutical Sciences, University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
e Department of Life Science, Faculty of Science, Okayama University of Science, 1-1 Ridai-cho, Kita-ku, Okayama 700-0005, Japan
f Department of Surgical Oncology, Osaka City University, Graduate School of Medicine, 1-4-3 Asahimachi, Abeno-ku, Osaka 545-8585, Japan
g Institute for Protein Research, Osaka University, 6-2-3 Furuedai, Suita, Osaka 565-0874, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
In the continuing study directed toward the development of peroxisome proliferator-activated receptor
Received 26 December 2012
Revised 2 February 2013
Accepted 4 February 2013
Available online 14 February 2013
gamma (hPPAR
c) agonist, we attempted to improve the water solubility of our previously developed
hPPAR -selective agonist 3, which is insufficiently soluble for practical use, by employing two strategies:
c
introducing substituents to reduce its molecular planarity and decreasing its hydrophobicity via replace-
ment of the adamantyl group with a heteroaromatic ring. The first approach proved ineffective, but the
second was productive. Here, we report the design and synthesis of a series of
acid-type hPPAR partial agonists with improved aqueous solubility. Among them, we selected (R)-7j,
which activates hPPAR to the extent of about 65% of the maximum observed with a full agonist, for fur-
ther evaluation. The ligand-binding mode and the reason for the partial-agonistic activity are discussed
based on X-ray-determined structure of the complex of hPPAR ligand-binding domain (LBD) and (R)-7j
a-benzyl phenylpropanoic
Keywords:
c
Human peroxisome proliferator-activated
receptor
PPAR gamma
c
c
Partial agonist
with previously reported ligand-LDB structures. Preliminal apoptotic effect of (R)-7j against human scir-
rhous gastric cancer cell line OCUM-2MD3 is also described.
Structural biology
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
reported to induce cell differentiation and apoptosis in malignant
tumor cells,11 and hPPAR
c
ligands may have potential as antican-
Peroxisome proliferator-activated receptor
c (hPPARc) is the
cer agents. We are interested in the potential application of
most extensively investigated subtype among the three hPPAR
subtypes, hPPAR , hPPARd and PPAR . It is expressed abundantly
in adipose tissue and macrophages,1 and is up-regulated in various
types of cancer cells.2,3 hPPAR
is a master regulator of fatty acid
and glucose homeostasis,4–6 but also has a range of other activi-
ties.7–10 Some hPPAR
ligands, such as pioglitazone (1) have been
hPPARc ligands to treat scirrhous gastric cancer, which has a poor
a
c
survival rate of less than 20%.12,13
We have extensively studied the structural design and synthesis
of subtype-selective hPPAR ligands based on phenylpropanoic acid
c
as a scaffold. We have successfully created a series of
phenylpropanoic acid-type hPPAR
-selective agonists,14,15
hPPAR
d dual agonist (TIPP-401),16,17 a hPPARd-selective agonist
(TIPP-204),18 and a hPPAR pan agonist (TIPP-703 (2)).19 Fur-
thermore, we recently reported hPPAR -selective agonist
(MEKT-1 (3)), which was designed on the basis of a comparative
a-substituted
c
a
a
a
adc
a
c
Abbreviations: NR, nuclear receptor; hPPAR, human peroxisome proliferator-
activated receptor; LBD, ligand-binding domain; PDB, protein data bank.
X-ray crystallographic analysis of 3 complexed with the hPPARc li-
⇑
Corresponding author. Tel.: +81 086 251 7930.
gand binding domain (LBD) (Fig. 1).20
E-mail address: miyachi@pharm.okayama-u.ac.jp (H. Miyachi).
Present address: International Institute for Advanced Studies 9-3, Kizugawadai,
Kizugawa City, Kyoto 619-0225, Japan.
Although 3 exhibited potent, hPPARc-selective in vitro transacti-
vation activity, its aqueous solubility is very poor. Therefore, for
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.02.003

2320
M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
1
2
3
Figure 1. The chemical structures of pioglitazone (1) and our hPPARc agonists 2 and 3.
in vivo evaluation studies, it was necessary to improve the solubility
of 3. Here, we report the design, synthesis and in vitro pharmacolog-
of 1 mL/min; solvent, MeCN/0.1% TFA = 3:1 v/v; detection at
254 nm or 296 nm).
ical evaluation of a series of
a-benzyl phenylpropanoic acid-type
hPPAR -selective partial agonists with improved aqueous solubil-
c
2.3. Thermodynamic aqueous solubility
ity, as candidate anti-scirrhous gastric cancer agents. We also
discuss the binding mode of these compounds and the reason why
they exhibit partial agonistic activity, based a comparison of the
Thermodynamic solubility determination was based on the
method of Avdeef and Testa.22 The concentration of sample solu-
tion was calculated using a previously determined calibration
curve, corrected for the dilution factor of the sample.
X-ray-determined structures of ligand complexes with the hPPAR
c
ligand binding site.
2.4. Transient transfection assays
2. Experimental design
2.1. Chemistry
The African green monkey kidney cell line CV-1 was used for
the transfection assay. CV-1 cells were seeded in 24-well plates
and cultured for 24 h. The transfection mixtures were added to
the cells and the plates were incubated for 5 h according to the
manufacturer’s instructions. After the transfection, incubation
was continued for an additional 40 h in the presence of the assay
compounds or reference compounds. Cell lysates were prepared
with a lysis buffer and used in the luciferase and bGAL assays.
Luciferase and bGAL activities were measured according to the
methods of Umesono et al.,23 with slight modifications. A substrate
reagent kit (Picagene, Toyo Ink) was used for the luciferase assay.
The synthesis of the present series of compounds is illustrated
in Scheme 1 (routes 1, 2). Compounds (R)-7a–(R)-7c and (R)-7e
were prepared according to reported methods, by amide-alkyl-
ation21 of the formyl derivative 6 with the corresponding benzam-
ides, followed by LiOOH hydrolysis (route 1).
Compounds (R)-7d and (R)-7f–(R)-7l were prepared from 6 by
an alternative method (route 2). Compound 6 was treated with
hydroxylamine HCl, and subsequent reduction with 10% Pd on car-
bon afforded aminomethylbenzene derivative 9 as the hydrochlo-
ric acid salt. Compound 9 was treated with the appropriate
benzoic acid derivatives in the presence of the condensing agent,
and subsequent hydrolysis with LiOOH afforded the desired
compounds.
2.5. Expression, purification, and crystallization
Human PPAR
c
LBD (residues 203–477) was prepared as previ-
LBD was expressed
ously described,24 that is, recombinant hPPAR
c
as an N-terminal His-tagged protein in Escherichia coli BL21 (DE3),
using the pET28a vector (Novagen). The protein was purified by
means of several steps of column chromatography (His-Trap HP
column (GE Healthcare), HiPrep 26/10 desalting column (GE
2.2. Hydrophobicity assessment with HPLC
Reversed-phase HPLC analyses were performed on an analytical
column (PegasilODS SP100 column (4.6 mm ꢀ 150 mm; flow rate
Scheme 1. Synthetic route to the present series of compounds. Reagents and conditions: (a) (1) BnBr, KHCO3, DMF, rt, overnight, quant.; (2) nPrI, K2CO3, DMF, 60 °C,
overnight, quant.; (3) NaBH4, EtOH, rt 2 h, 91%; (4) PBr3, diethylether, 0 °C, 1.5 h, 52%; (b) (1) (R)-3-(3-phenylpropanoyl)-4-benzyloxazolidin-2-one, LiHMDS, dehydrated THF,
ꢁ50 to 0 °C, 2 h, 64%; (2) H2, 10% Pd-C, AcOEt, rt, 3 h, 55%; (3) BH3–THF, dehydrated THF, 0 °C, overnight; (4) activated MnO2, CH2Cl2, rt, overnight, 65% (2 steps); (c) (1)
benzamide derivative, triethylsilane, TFA, toluene, reflux, 48 h, 97%; (2) LiOH–H2O, 30% H2O2, THF/H2O = 4:1 (v/v), 0 °C, 2.5 h then rt 3 h, 86%; (d) hydroxylamine HCl,
pyridine, ethanol, reflux, 24 h, 93%; (e) H2, 10% Pd-C, ethanol–HCl, rt, 3 h, 96%; (f) (1) benzoic acid derivative, diethyl cyanophosphonate, triethylamine, DMF, rt, 12–24 h, 46–
82%; (2) LiOH–H2O, 30% H2O2, THF/H2O = 4:1 (v/v), 0 °C, 2.5 h then rt 3 h, 51–94%.

M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
2321
Healthcare), Benzamidine FF column (GE Healthcare), HisTrap HP
column (GE Healthcare), HiTrap Q HP column (GE Healthcare), Hi-
biphenyl derivative (R)-7a as a lead compound. The poor aqueous
solubility of 3 might be attributed primarily to the presence of an
adamantyl substituent at the hydrophobic tail part of the molecule,
even though this substituent is one of the critical determinants for
Prep 26/10 desalting column (GE Healthcare)). Apo PPAR
c LBD
crystals were soaked in ligand solution for two or three weeks prior
to crystallography.
potent hPPAR
derivative (R)-7a exhibited decreased activity compared with 3,
it still showed hPPAR agonistic activity at sub-micromolar con-
c-selective agonistic activity. Although biphenyl
2.6. Diffraction data collection
c
centration (EC50 = 86 nM in our assay system).
The diffraction data for hPPAR
c
LBD complex with (R)-7j were
We have previously reported biphenylcarboxylic acid-type
hPPARd partial agonists with improved aqueous solubility.33 In
that case, the improvement of solubility was attributed to two fac-
tors, that is, disruption of molecular planarity by the introduction
of a substituent at the ortho-position of the attached benzene ring,
and decreased hydrophobicity due to the introduction of a hetero-
aromatic ring. We also applied the same strategies in the present
work, and the results are summarized in Table 1.
We evaluated the aqueous solubility of the compounds by calcu-
lating ClogP values34 and we also determined the retention times of
the compounds in reversed-phase high-performance liquid chro-
matography (RP-HPLC). In the case of adamantyl derivative 3, the
ClogP value is extremely high, more than 9, and the RP-HPLC reten-
tion time is about 30 min. These data clearly indicated that 3 is ex-
tremely hydrophobic, being even more hydrophobic than typical
steroid drugs. However, in the case of the phenyl derivative (R)-
7a, ClogP is about 7.1, so that this compound is about 100 times
more hydrophilic than 3. In accordance with this, compound (R)-
7a was eluted much faster than 3, with a retention time of 7.9 min.
The ortho-substituted derivatives ((R)-7b, (R)-7c) exhibited
collected on BL38B1 at SPring-8 (Harima, Japan) and were pro-
cessed using HKL-2000.25
2.7. Structure analysis and refinement
The structure of hPPARa LBD complex with (R)-7j was solved by
the molecular-replacement method with the program CNS26 using
the previously reported structures as probes. The correctly posi-
tioned molecules were refined with CNS and O.27 The initial atomic
model of (R)-7j was built using MOE (Ryoka Systems Inc.) and
topology and parameter files for the refinement were generated
by the HIC-Up server.28
The structure of hPPAR
replacement method using Molrep. Apo PPAR
c
LBD was solved by the molecular
LBD structure
c
(PDB code 1PRG) was used as a search model and refined with Ref-
mac5. The crystallographic data and data collection statistics of all
the crystals are provided in Table 1.
2.8. Adipogenesis assay
comparable ((R)-7b) or somewhat increased ((R)-7c) hPPARc ago-
The adipogenesis assay was performed as previously described.29
Mouse adipogenic fibroblast cells, 3T3-L1, were cultured in DMEM
with 10% FBS. After the cells became confluent, they were pretreated
nistic activity. However, their aqueous solubility was not im-
proved; indeed in the case of (R)-7b, it was decreased as
compared to the non-substituted derivative (R)-7a. As expected,
introduction of a small fluorine atom at the meta or para position
((R)-7d, (R)-7e) of the distal benzene ring did not significantly af-
fect the in vitro biological activity or the aqueous solubility.
It is interesting to note that thiophen-2-yl phenyl derivative
(R)-7f exhibited somewhat greater aqueous solubility with reten-
with 1
thine (IBMX), and 5
l
M dexamethasone (Dex), 0.5 mM 3-isobutyl-1-methylxan-
g/mL insulin to initiate adipogenesis for 2 days.
l
Various concentrations of ligands were added to the cells, and after
6 days, the cells were stained with Oil Red O. The degree of adipo-
genesis was quantitatively measured in terms of absorbance at
OD550. The error bars are the standard deviations (SD).
tion of the hPPARc agonistic activity, as compared to the phenyl
derivative (R)-7a. Although the thiophene ring, a well-known bio-
isostere of the benzene ring, is less hydrophobic than the benzene
ring, the hydrophobicity of (R)-7f was increased. The reason for
this is unclear. In any event, unsymmetrization of the biphenyl
moiety of (R)-7a by the introduction of a substituent at the
ortho-position of the benzene ring or by replacement of one of
the benzene rings with an bioisostere was not very effective in this
case. Therefore, we next tried to substitute a heterocyclic ring (pyr-
idine, ((R)-7g–(R)-7i), pyrimidine ((R)-7j, (R)-7l), pyrazine ((R)-
7k)) at the distal benzene ring. The results are included in Table 1.
The position of the nitrogen atom of the monoazine is important
2.9. Cell lines
A human scirrhous gastric cancer cell line, OCUM-2MD330 was
cultured in DMEM/F12 (Ham) (1:1) (Invitrogen, Carlsbad, CA). A
normal human fibroblast cell line, OUMS-24,31 was cultured in
DMEM (Nissui). All media were supplemented with 10% fetal bo-
vine serum (Invitrogen), 100
Japan) and 0.5 g/mL Fungizone (Invitrogen).
lg/mL kanamycin (Meiji Seika, Tokyo,
l
2.10. Apoptosis assay
for potent hPPARc agonistic activity, that is, the hPPARc agonistic
activity decreased in the order of 2-pyridyl ((R)-7g) > 3-pyridyl
((R)-7h) > 4-pyridyl ((R)-7i), and pyrimidin-2-yl ((R)-7j) > pyra-
din-2-yl ((R)-7h) > pyrimidin-5-yl ((R)-7i). These results indicated
that the distal nitrogen atom interacts unfavorably with the
The apoptosis assay was performed as previously described.32
Cells were inoculated into flat-bottomed 6-well plates, incubated
for 24 h and then treated with test compound. Seventy-two hours
later, 1 lg/mL Hoechst 33342 (Invitrogen) and 5 lg/mL propidium
hPPARc LBD, causing a considerable decrease of the activity.
iodide (Sigma) were added to the medium and the cells were incu-
bated in the dark for 20 min. All of the cells were collected on a glass
slide and evaluated with a fluorescence microscope. Cells with frag-
mented or shrunken nuclei were counted as apoptotic cells.
As for aqueous solubility, the introduction of a pyridine ring in-
stead of the benzene ring as the distal substituent improved the
aqueous solubility by 10-fold, as judged from ClogP (7.18 to 6.04
and 5.83). In accordance with this, (R)-7g–(R)-7i (retention times:
3.6, 2.3 and 1.8 min, respectively) were eluted much faster than the
phenyl analogue (R)-7a (retention time: 7.9 min). These com-
pounds exhibited similar ClogP values, whereas their retention
times were somewhat different, that is, the compounds eluted fas-
ter in the order of (R)-7i (4-pyridyl) > (R)-7h (3-pyridyl) > (R)-7g
(2-pyridyl). Thus, it appears that these two hydrophobic parame-
ters do not necessarily correlate well.
3. Results and discussion
3.1. Structure–activity and structure–solubility relationships
In order to create phenylpropanoic acid-type PPAR
c-selective
agonists with improved aqueous solubility, we focused on the

2322
M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
Table 1
hPPAR
c
transactivation activities, ClogP values and retention times (rt) on reversed-phase HPLC of the present series of derivatives
O
O
N
H
OH
H
R1
O
a
a
No.
R
EC50
Efficacyb ClogPc rtd (min)
Aqueous
No.
R
EC50
Efficacyb ClogPc rtd (min)
Aqueous
(nM)
hPPAR
3.60
solubilityf)
(nM)
solubilityf)
(lg/mL)
(lg/mL)
c
hPPARc
3
NTe
9.58
7.18
30.5 0.02 <1.00
7.89 0.01 <1.00
7a
85.9
73.8
95.7
7g
7h
186.1
844
113
100
6.04
5.83
3.57 0.01
2.31 0.01
17.0
69.5
7b
7c
72.7
95.0
7.38
7.37
9.71 0.01 <1.00
7.66 0.01 <1.00
49.5
28.7
7i
7j
1688
34.6
108
5.83
5.12
1.78 0.01
4.25 0.01
34.1
301
7d
7e
99.7
7.37
8.10 0.01 <1.00
65.0
102.6
65.2
104
7.37
7.37
7.82 0.02 <1.00
7.42 0.01 <1.00
7k
7l
713.9
3095
58.2
82.7
5.12
4.91
3.845 0.01 32.2
7f
76.0
3.23 0.01
190
a,b
EC50 values and % efficacy are given relative to the positive control, rosiglitazone for PPARc.
CLogP values were estimated with ChemDraw Ultra version 10.0.
c
d
e
f
PegasilODS SP100 column (4.6 mm ꢀ 150 mm; flow rate of 1 mL/min; solvent, MeCN: 0.1% TFA = 3:1 v/v; detection at 254 nm or 296 nm).
NT means not tested.
Aqueous solubility in phosphate buffer (pH = 7.2–7.4).
In the case of (R)-7j–(R)-7l, the position of the nitrogen atom is
also important for potency towards hPPAR . The pyrimidin-2-yl
of about 65% of the maximum activity observed with the full-ago-
nist pioglitazone (1).
c
derivative ((R)-7j) exhibited potent activity, comparable to that
of the phenyl analog ((R)-7a). However, others, that is, the pyra-
zin-2-yl ((R)-7k) and pyrimidin-5-yl ((R)-7l) derivatives, exhibited
3.2. Partial-agonistic activity of (R)-7j
decreased hPPAR
meta position nitrogen atoms interact unfavorably with hPPAR
LBD, thereby decreasing the activity.
c
agonistic activity. These results indicated that
To characterize the hPPAR
compared the ability of 2 (an hPPAR
stimulate adipocyte differentiation of murine preadipocyte 3T3-
L1 cells (Fig. 2). It is well known that PPAR agonists promote the
conversion of a variety of preadipocyte and stem cell lines into ma-
ture adipocytes.35 Both compounds in the concentration range from
1 to 1000 nM stimulated adipocyte differentiation and induced tri-
glyceride accumulation, as indicated by Oil Red O staining. Com-
c
partial-agonistic profile of (R)-7j, we
c
c
full-agonist), and (R)-7j to
As for aqueous solubility, (R)-7j–(R)-7l exhibited about 100-
fold increased aqueous solubility, as judged from ClogP (from
7.18 for (R)-7a to 5.12, 5.12 and 4.91, respectively). In accordance
with this, these diazines were eluted more quickly than the phenyl
analogue (R)-7a. However, they were eluted more slowly than the
pyridine derivatives (R)-7g–(R)-7i. These results indicate that
introduction of hydrophilic aromatics, such as a pyridine ring or
pyrimidine ring, is an effective strategy to improve the aqueous
solubility of the present series of compounds.
We then assessed the exact thermodynamic solubility of these
compounds. As expected, all the biphenyl derivatives ((R)-7a–
(R)-7f) exhibited very poor aqueous solubility (<1
the introduction of a nitrogen atom at the distal benzene ring im-
proved the solubility to 17.0 g/mL ((R)-7g) and 301 g/mL ((R)-
c
pound
2 exhibited dose-dependent adipocyte differentiation-
promoting ability, providing the maximum absorption intensity
(OD550 = 0.2 in these experiments). On the other hand, (R)-7j stim-
ulated adipocyte differentiation at the lowest concentration of
1 nM, but dose-dependency was not maintained at high concentra-
tion and the absorption intensity was about 60% of the maximum at
every concentration tested. These results indicate that (R)-7j is a
lg/mL), but
hPPARc partial-agonist.
l
l
7j). The number of introduced nitrogen atoms was correlated with
3.3. Structural features of the complex of (R)-7j with hPPARc
the thermodynamic solubility. We selected compound (R)-7j for
ligand-binding domain (LBD)
further evaluation, based on its activity towards hPPAR
aqueous solubility. The intrinsic activity of (R)-7j suggests that it is
a partial agonist, with the ability to activate hPPAR to the extent
c and good
To understand the structural basis for the activity of (R)-7j, we
examined the three-dimensional structure of (R)-7j complexed
c

M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
2323
with the hPPARc LBD by means of X-ray crystallographic analysis.
The resulting three-dimensional structures of hPPAR
complex are illustrated in Figure 3A–G. The crystallographic data
c
LBDꢁ(R)-7j
and refinement statistics are summarized in Table 2.
Like the reported structure of the hPPAR
hPPAR LBD has a three-layered sandwich folded structure com-
prising mainly -helices, and the overall interaction mode with
c
LBDꢁ3 complex,
c
a
the ligand is basically conserved.
The pyrimidin-2-yl derivative (R)-7j exhibited somewhat weak-
er hPPAR
c
agonistic activity than the adamantyl derivative 3, and
LBDꢁ(R)-7j complex
the X-ray crystallographic analysis of hPPAR
c
clearly revealed the reason for this difference. The pyrimidin-2-yl
ring and adamantyl ring are hosted in the same binding pocket,
that is, the large Y3 arm36 cavity composed of the side chains of
amino acid residues Leu255, Glu259, Ile262, Phe264, Arg280 and
Ile281 (Fig. 3D and E). The planar pyrimidin-2-yl ring would inter-
act less effectively with this large hydrophobic cavity, as compared
to the three-dimensionally expanded hydrophobic adamantyl ring,
especially as regards the hydrophobic interactions with Ile262,
Phe264, and Ile 281, which are positioned on both sides of the pyr-
imidin-2-yl ring of (R)-7j (Fig. 3F and G).
Figure 2. Adipogenesis-inducing activity of 3 and (R)-7j. After pretreatment with
M Dex, 0.5 mM IBMX, and 5 g/mL insulin, 3T3-L1 cells were incubated with the
indicated concentrations of 3 and (R)-7j, respectively. Oil red O-stained cells are
1
l
l
shown at the top and are quantified at the bottom by measurement of OD550 ( SD).
Figure 3. (A–G) Crystal structures of hPPAR
Protein is represented as a blue ribbon model and (R)-7j is depicted as a magenta cylinder model. The numbering of the second structure is also depicted. The nomenclature of
the helices is based on the RXR- crystal structure. (B) The superimposed structures of 3 and (R)-7j complexed with hPPAR LBD. (C) Whole structure of hPPAR LBD–3
c LBD–3 complex (PDB: 2ZNO) and hPPARc LBD–(R)-7j complex (PDB: 3VSO). (A) Whole structure of hPPARcLBD–(R)-7j complex.
a
c
c
complex. Protein is represented as a white ribbon model and 3 is depicted as a green cylinder model. (D) Zoomed view of the binding mode of the hydrophobic tail part of 3 in
the Y3 arm. The side chains of amino acids of the Y3 arm are depicted as blue cylinder models. (E) Zoomed view of the binding mode of the hydrophobic tail part of (R)-7j in
the Y3 arm. (F) Zoomed view of the alinement of the hydrophobic tail part of 3 and (R)-7j. (G) image F rotated by 90°.

2324
M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
Table 2
folds are well matched in these two complexes, and the positions
Crystallographic data and refinement statistics
and the directions of the side chains of Lys301 and Glu471 overlap
well in these structures. The root mean square deviations (RSMDs)
of these two amino acid residues are 0.71 and 0.44 Å, respectively
Crystal
PPARg-7j
Data collection
Space group
Unit cell constants
a (Å)
b (Å)
c (Å)
(Fig. 4E, F).42 These results indicate that the two hPPAR
c full ago-
C2
nists will be able to recruit coactivator complexes with almost
the same efficacy, eliciting full agonistic activity.
93.35
61.71
118.98
102.79
On the other hand, the overall structures of hPPAR
plexed with hPPAR partial agonists, including (R)-7j, are some-
what different. The frameworks of hPPAR LBD complexed with
partial agonists ((R)-7j, LT127,43
c LBD com-
c
b (°)
c
Wavelength (Å)
Resolutiona (Å)
(Highest shell)
Measured reflections
Unique reflections
Completeness
I/r(I)
Redundancy
Rmerge
Refinement
1
50.0–1.97
2.04–1.97
172342(16510)
46474(4586)
99.2(99.7)
17.2(3.7)
3.7(3.6)
several representative hPPAR
c
phenoxyacetic acid derivative,44 GW-0072,45 pyrazole derivative,46
and GQ-1647) are depicted in Figure 5A–F, and a superposition of
the six complexes is shown in Figure 5G. The folds of the upper half
of the hPPAR
lixes, are well matched in these complexes. However, the folds of
the lower half of the hPPAR
LBD, composed of H20, H3, H6,
c LBD, composed of the H1, H2, H4, H5, H8 and H9 he-
b
4.2(40.1)
c
H10–H12 helixes and b1–b3 sheets, are somewhat different. The
position of Lys301 in the H3 helix varies somewhat
(RMSD = 1.10 Å, as compared to 0.71 Å for full agonists) in these
structures. However, the side chain of Glu471 in the H12 helix
faces in various directions in these complexes, and the RSMD is
much larger (1.89 Å as compared to 0.44 Å for full agonists). These
substantial differences in structure might be one of the driving
forces behind the partial agonistic activity of these compounds,
including (R)-7j.
Resolution (Å)
30.0–2.0
21.9/25.7
d
Rworkc/Rfree (%)
Number of atoms
Protein
Water
4184
117
38
Ligand
Average B-factor (Å2)
Protein
41.3
36.7
57
Water
Ligand
r.m.s.d.
Bond Lengths (Å)
Angles (°)
PDB code
To better understand the key differences in the binding of
hPPARc full agonists and hPPARc partial agonists, we focused on
the binding mode of these agonists to Tyr473 in the H12 helix.
0.007
1.0
3VSO
a
As already reported, Tyr473 is believed to play a critical role in
Values in parentheses indicate statistics for the highest resolution shells.
b
the full agonist-mediated stabilization of the LBD of hPPAR
c
Rmerge
=
RhklRi|Ii(hkl) ꢁ hI(hkl)i|/RhklRiIi(hkl), where hI(hkl)i is the mean I(hkl)
over symmetry-equivalent reflections.
(Fig. 6A–C). In the case of the full agonist rosiglitazone, the side
chain phenol oxygen of Tyr473 forms a tight hydrogen bond with
the nitrogen atom of the thiazolidine-2,4-dione ring. Furthermore,
the side chain phenol oxygen of Tyr473 has a hydrogen bonding
interaction with the carbonyl oxygen atom of the thiazolidine-
2,4-dione ring. These hydrogen bonding interactions tightly fix
Tyr473 in the H12 helix, forming a section of the tranccriptional
c
Rwork
=
Rhkl|Fobs ꢁ Fcalc|/Rhkl|Fobs|, where Fobs and Fcalc are the observed and
calculated structure factors, respectively.
d
Rfree was calculated using 5% of the total reflections, which were chosen ran-
domly and omitted from the refinement.
It is interesting to note Glu259 and Arg280 show a hydrogen
bonding interaction to form a ‘salt bridge’37 that determines the
coactivator-binding surface of the PPARc LBD Figure 6A.
depth of the Y3 arm. The rank order of the decrease in hPPARc ago-
In the case of the partial agonist N-[1-(4-fluorophenyl)-3-(2-the-
nyl)-1H-pyrazole-5-yl]-3,5-bis- (trifluoromethyl)benzenesulfonamide
(PDB: 2GOH), the 4-fluorophenyl ring is located close to the H12 he-
lix, but it has no hydrophobic or hydrogen bonding interaction with
this helix (Fig. 6C). Therefore, there is no significant ligand-mediated
interaction to fix Tyr473 in the H12 helix. The absence of interaction
(or the presence of weak interaction) with Tyr473 in the H12 helix
could provide the structural basis for the partial agonist functional-
ity of not only this compound, but also the other neutral-and carbox-
nistic activity of 4-pyridyl ((R)-7i) > 3-pyridyl ((R)-7h) > 2-pyridyl
((R)-7g) derivatives might be attributed to unfavorable interaction
between the nitrogen atom of the distal heteroaromatic ring and
the anionic side chain of Glu259.
3.4. Structural basis of the partial agonism of (R)-7j towards
hPPAR
c
The qualitative characteristics of corepressor dissociation and
coactivator recruitment are reported to be regulated differentially
by the three-dimensional structure around the C-terminal H12 he-
lix and the surrounding helixes of the nuclear receptor.38 In the
case of the binding of a full agonist, such as rosiglitazone, the
H12 helix forms a part of the coactivator-binding surface, along
with the H3 helix and H5 helix.39 In this case, the positions and
directions of the side chains of two distinct hydrophilic amino
acids, Lys301 (in H3 helix) and Glu471 (in H12 helix), are reported
to be critically important, because these amino acids clamp the
LXXLL motif (L, leucine; X, any amino acid) of the coactivator,
and dock it appropriately to express full-agonistic activity.39 There-
fore, we first compared the three-dimensional structures of two
ylic acid-type hPPAR
c partial agonists shown in Figures 5B–F 6B.
In the case of -benzylphenylpropanoic acid (R)-7j, the acidic
a
carboxylic acid moiety hydrogen bonds with the side chain phenol
oxygen of Tyr473. However, the distance between these two func-
tionalities is longer than that in the case of hPPAR
with the thiazolidine-2,4-dione-type hPPAR full agonist, rosiglit-
azone. Furthermore, there is only a single hydrogen bonding inter-
action between the acidic moiety of (R)-7j and Tyr473 in the H12
helix. Considering these two factors, Tyr473 in the H12 helix is
thought to be fixed less effectively when (R)-7j is bound to
c LBD complexed
c
hPPARc-LBD, resulting in partial agonistic activity Figure 6B.
3.5. Preliminal apoptotic effect of (R)-7j on scirrhous gastric
cancer cell line
PPAR
with hPPAR
The main frameworks of hPPAR
c
full agonists, rosiglitazone and farglitazar,40 complexed
c
LBD41 (Fig. 4A–F).
c
LBD complexed with rosiglit-
In order to investigate the potential of hPPARc partial agonist
azone and farglitazar are depicted in Figure 4A, B, and a superposi-
tion of these two complexes is shown in Figure 4C. The overall
(R)-7j as an anti-scirrhous gastric cancer agent, we preliminally
compared the apoptosis-inducing effect of (R)-7j and 3, together

M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
2325
Figure 4. (A, B) Crystal structures of hPPAR
superimposed structures of hPPAR LBD–rosiglitazone complex and hPPAR
highlighted as cylinder models in magenta (rosiglitazone complex) and cyan (farglitazar complex). The N-terminal AF2 region (H12 helix) is highlighted as a light green
dotted box. (D) Chemical structures of rosiglitazone and farglitazar. (E) Zoomed view of the superimposed Lys301 of hPPAR LBD–rosiglitazone complex and hPPAR LBD–
farglitazar complex. Lys301 is highlighted as a cylinder model in magenta (rosiglitazone complex) or blue (farglitazar). (F) Zoomed view of the superimposed Glu473 of
hPPAR LBD–rosiglitazone complex and hPPAR LBD–farglitazar complex. Glu473 is highlighted as a cylinder model in magenta (rosiglitazone complex) or blue (farglitazar).
c
LBD–rosiglitazone (full agonist) complex (PDB: 2PRG) and hPPAR
c
LBD–farglitazar (full agonist) complex (PDB: 1FM9). (C) The
c
c LBD–farglitazar. Both ligands are depicted as cylinder models, and Lys301 and Glu473 are
c
c
c
c
with the hPPAR
c
full agonist troglitazone, on human scirrhous gas-
7j also dose-dependently enhanced apoptosis in the concentration
range of 1–100 M. The potency was equal to or somewhat greater
than that of the hPPAR full-agonist troglitazone (based on the
activity at 10 M). Furthermore, (R)-7j shows high selectivity for
cancer cells, comparable to that of troglitazone. These results indi-
cated that partial activation of PPAR might sufficient to induce
apoptosis of human scirrhous gastric cancer cells.
In conclusion, we have created the novel -benzyl phenylprop-
anoic acid-type hPPAR partial agonist (R)-7j, which shows in-
creased aqueous solubility as a result of replacement of the
adamantyl group of 3 with a pyrimidin-2-yl group. It also selec-
tively induced apoptosis of OCUM-2MD3 human scirrhous gastric
cancer cells. In vivo pharmacological evaluation of 7j as anti-scir-
rhous gastric cancer agent is under way.
tric cancer cell line OCUM-2MD3. Troglitazone was used as a posi-
tive control, based on reports that it suppresses the growth of
gastric cancer cell lines, such as MKV45 cells,48 SNU-216 and
SNU-668 cells.49
OCUM-2MD3 cells and OUMS-24 (a normal human fibroblast
cell line) cells were incubated for 24 h and then treated with the
indicated concentrations of test compounds. After 72 h, Hoechst
33342 and propidium iodide were added to the medium, and the
cells were incubated in the dark for 20 min. All the cells were then
collected on a glass slide, and evaluated under a fluorescence
microscope. Cells with fragmented and/or shrunken nuclei were
counted as apoptotic cells. The results, expressed as % increase of
apoptotic cells versus the DMSO control, are illustrated in Figure 7.
l
c
l
c
a
c
Troglitazone (hPPAR
apoptosis in the concentration range of 1–100
c
full agonist) dose-dependently enhanced
M. Thus, troglitaz-
4. Experimental section
l
one induces apoptosis of human scirrhous gastric cancer cell line
OCUM-2MD3, as well as the other gastric cancer cell lines,
MKV45, SNU-216 and SNU-668,48,49 while it exhibited little effect
on normal cells. The finding that troglitazone selectively induced
apoptosis of human scirrhous gastric cancer cells is important,
even though the reason is unknown. One possibility is that differ-
4.1. Chemistry: general methods
Melting points were determined with a Yanagimoto hot-stage
melting point apparatus and are uncorrected. 1H NMR and 13C
NMR spectra were recorded on a Varian VNMRS-400 (400 MHz)
SC-NMR spectrometer. Proton chemical shifts were referenced to
TMS as an internal standard. FAB-MS was carried out with a
VG70-SE. Chiral HPLC analyses were performed on an analytical
column (CHIRALPAK IA column (4.6 ꢀ 150 mm; detection at
254 nm). HPLC analyses were performed on an analytical column
(PegasilODS SP100 column (4.6 ꢀ 150 mm; flow rate of 1 mL/
min; solvent, MeCN/0.1% TFA = 3:1 v/v; detection at 254–
296 nm). All optical rotation data was recorded at 25 °C on a Jasco
P-2000 polarimeter with a 50 mm cell (concentration reported as
g/100 mL).
ent amounts of PPAR
The hPPAR full agonist 3 also dose-dependently enhanced
apoptosis in the concentration range of 1–100 M, and it was more
potent than troglitazone (10 M 3 effectively induced apoptosis of
OCUM-2MD3 cells). However, 100 M 3 induced apoptosis of both
c may be expressed in the cell lines.
c
l
l
l
OCUM-2MD3 and OCUM-24 cell lines. Therefore, 3 may show
poorer selectivity for human scirrhous gastric cancer cells, as com-
pared to troglitazone.
Finally, we found that hPPARc partial agonist (R)-7j also effec-
tively induced apoptosis of OCUM-2MD3 cells. Treatment of (R)-

2326
M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
Figure 5. (A–F) Crystal structures of hPPAR
LBD–LT127 complex (PDB: 2I4Z). (C) Crystal structure of hPPAR
c
LBD–partial agonist complexes. (A) Crystal structure of hPPAR
LBD–(2S)-2-(4-chlorophenoxy)-3-phenylpropanoic acid complex (PDB: 3CDP). (D) Crystal structure of
LBD–GW-0072 complex (PDB: 4PRG). (E) Crystal structure of hPPAR LBD–N-[1-(4-fluorophenyl)-3-(2-thenyl)-1H-pyrazole-5-yl]-3,5-bis(trifluoromethyl)benzene-
sulfonamide complex (PDB: 2GOH). (F) Crystal structure of hPPAR LBD–GQ-16 complex (PDB: 3T03). (G) The superimposed structures of hPPAR LBD–partial agonist
complexes. All ligands are depicted as cylinder models, and Lys301 and Glu473 are highlighted as green cylinder models. The N-terminal AF2 region (H12 helix) is highlighted
as a light green dotted box. (H) Zoomed view of the superimposed Lys301 of hPPAR LBD–partial agonist complexes. All Lys301 residus are highlighted as a green cylinder
model. (I) Zoomed view of the superimposed Glu473 of hPPAR LBD–partial agonist complexes. All Glu473 residues are also highlighted as a green cylinder model.
c LBD–7j complex (PDB: 3VSO). (B) Crystal structure of hPPARc
c
hPPAR
c
c
c
c
c
c
4.2. (R)-2-(3-([1,10-Biphenyl]-4-ylcarboxamidomethyl)-4-
propoxybenzyl)-3-phenylpropanoic acid ((R)-7a)
CDCl3) d 7.74 (d, J = 8.4 Hz, 2H), 7.58 (t, J = 8.6 Hz, 4H), 7.45 (t,
J = 7.4 Hz, 2H), 7.37 (t, J = 7.4 Hz, 1H), 7.29–7.15 (m, 10H), 6.88–
6.85 (m, 2H), 6.81 (d, J = 8.4 Hz, 1H), 6.61 (t, J = 5.2 Hz, 1H), 4.69
(dd, J = 14.4, 5.6 Hz, 1H), 4.64–4.55 (m, 2H), 4.41–4.35 (m, 1H),
4.00–3.93 (m, 2H), 3.90 (dd, J = 8.8, 2.4 Hz, 1H), 3.74 (t, J = 8.2 Hz,
1H), 3.08 (dd, J = 13.4, 9.0 Hz, 1H), 2.98 (dd, J = 13.4, 9.0 Hz, 1H),
2.91–2.78 (m, 3H), 2.51 (dd, J = 13.4, 8.6 Hz, 1H), 1.87–1.78 (m,
2H), 1.04 (t, J = 7.4 Hz, 3H); MS (FAB) 667 (M+H)+.
This intermediate (157 mg, 0.24 mmol) was dissolved in 16 mL
of tetrahydrofuran and 4 mL of water under an argon atmosphere
with ice-cooling. To this solution was added 30% aqueous hydro-
gen peroxide (0.27 mL, 2.40 mmol). Then a solution of lithium
hydroxide monohydrate (40 mg, 0.96 mmol) in water (1 mL) was
added, and the mixture was stirred for 3.5 h at 0 °C. An aqueous
A mixture of 5-((R)-2-benzyl-3-((S)-4-benzyl-2-oxo-oxazolidin-3-
yl)-3-oxopropyl)-2-propoxy-benzaldehyde (6) (250 mg, 0.51 mmol),
4-phenylbenzamide (300 mg, 1.54 mmol), triethylsilane (0.25 mL,
1.54 mmol), trifluoroacetic acid (0.12 mL, 1.54 mmol) and 40 mL
of dehydrated toluene was refluxed for 48 h under an argon atmo-
sphere. The reaction mixture was poured into water, and the whole
was extracted with ethyl acetate. The extract was washed with
brine, dried over anhydrous magnesium sulfate and concentrated.
The residue was purified by silica gel column chromatography (elu-
ant; n-hexane/ethyl acetate = 1:1 v/v) to afford 162 mg (47%) of the
intermediate derivative as a pale yellow solid. 1H NMR (400 MHz,

M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
2327
Figure 6. (A–C) Zoomed views of the binding mode of the AF2 region (H12 helix) of hPPAR
c
LBD–rosiglitazone. (A) Zoomed view of hPPAR
c
LBD–rosiglitazone (full agonist)
complex. (B) Zoomed view of hPPAR
c
LBD–7j (partial agonist) complex. (C) Zoomed view of hPPAR
c
LBD–N-[1-(4-fluorophenyl)-3-(2-thenyl)-1H-pyrazole-5-yl]-3,5-
bis(trifluoromethyl)benzenesulfonamide (partial agonist) complex.
100
2MD3 --> 24
24 --> 2MD3
75
50
25
0
cont
0.1
1
10
100
cont
0.1
1
10
100
cont
0.1
1
10
100
Troglitazone
3
7j
[Compound
(µM)]
Figure 7. Apoptosis-inducing activity of troglitazone, 3 and (R)-7j on OCUM-2MD3 and OUMS-24 cells.
solution of sodium sulfite (1.00 g/6 mL) was further added and stir-
3.02–2.86 (m, 3H), 2.80–2.73 (m, 2H), 1.88–1.79 (m, 2H), 1.05 (t,
J = 7.4 Hz, 3H); 13C NMR (300 MHz, CDCl3) d 178.85, 167.07,
155.74, 144.11, 140.01, 138.92, 133.25, 130.90, 130.46, 129.18,
128.91, 128.87, 128.42, 127.92, 127.44, 127.16, 126.39, 125.94,
111.12, 69.50, 49.42, 40.31, 37.80, 36.92, 22.74, 10.74; MS (FAB)
ring was continued for 15 min. The reaction mixture was acidified
with 10% HCl, and then extracted with ethyl acetate. The extract
was washed with brine, dried over anhydrous magnesium sulfate
and concentrated. The residue was purified by silica gel column
chromatography (eluant; n-hexane/ethyl acetate = 1:1 v/v) to af-
ford 100 mg (82%) of the title compound as a white amorphous so-
lid. 1H NMR (400 MHz, CDCl3) d 7.80 (d, J = 8.4 Hz, 2H), 7.63–7.57
(m, 4H), 7.45 (t, J = 7.4 Hz, 2H), 7.37 (t, J = 7.4 Hz, 1H), 7.27–7.23
(m, 2H), 7.04 (dd, J = 8.4, 2.4 Hz, 1H), 6.81 (t, J = 5.8 Hz, 1H), 6.77
(d, J = 8.0 Hz, 1H), 4.65–4.55 (m, 2H), 3.95 (t, J = 6.4 Hz, 2H),
508 (M+H)+; calcd for C33H34NO4 508.2487, found 508.2503; [
a]
D
ꢁ8.7° (c 0.25, CH3CN). The enantiomeric excess (ee) of this com-
pound was determined to be >98% by chiral HPLC, using a
4.6 ꢀ 250 mm CHIRALPAK IA column (Diacel Chemical Industries,
Ltd), eluting at a flow rate of 0.5 mL/min with 2-propanol/n-hex-
ane/TFA (1:2:0.03 v/v/v): retention time 17.4 min (the S enantio-

2328
M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
mer has a retention time of 15.4 min); HPLC purity was estimated
to be 99.8% by means of reversed-phase HPLC, using a Pegasil ODS
sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted on a
Shimadzu HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as the elu-
ant and detection at 254 nm.
1H NMR (400 MHz, CDCl3) d 7.80 (d, J = 8.0 Hz, 2H), 7.57–7.52
(m, 4H), 7.27–7.23 (m, 2H), 7.20–7.11 (m, 6H), 7.04 (dd, J = 8.2,
2.2 Hz, 1H), 6.80–6.76 (m, 2H), 4.65–4.56 (m, 2H), 3.95 (t,
J = 6.4 Hz, 2H), 3.01–2.87 (m, 3H), 2.81–2.73 (m, 2H), 1.88–1.79
(m, 2H), 1.05 (t, J = 7.4 Hz, 3H); 13C NMR (300 MHz, CDCl3) d
179.13, 166.92, 164.42, 161.15, 155.57, 143.07, 138.82, 136.15,
133.27, 130.81, 130.45, 129.20, 128.90, 128.84, 128.74, 128.45,
127.52, 127.01, 126.44, 125.91, 115.95, 115.67, 111.12, 69.50,
49.35, 40.31, 37.73, 36.89, 22.74, 10.77; 13C NMR (300 MHz, CDCl3)
d 179.13, 166.92, 164.42, 161.15, 155.57, 143.07, 138.82, 136.15,
133.27, 130.81, 130.45, 129.20, 128.90, 128.84, 128.74, 128.45,
127.52, 127.01, 126.44, 125.91, 115.95, 115.67, 111.12, 69.50,
49.35, 40.31, 37.73, 36.89, 22.74, 10.77;MS (FAB) 526 (M+H)+;
4.3. (R)-2-Benzyl-3-(3-((20-methyl-[1,10-biphenyl]-4-
ylcarboxamido)methyl)-4-propoxy-phenyl)propanoic acid
((R)-7b)
This compound was prepared by means of a procedure similar
to that used for 7a. 1H NMR (400 MHz, CDCl3)
d 7.78 (d,
J = 8.4 Hz, 2H), 7.35 (d, J = 8.4 Hz, 2H), 7.28–7.22 (m, 5H), 7.20–
7.17 (m, 5H), 7.04 (dd, J = 8.4, 2.2 Hz, 1H), 6.81–6.76 (m, 2H),
4.65–4.56 (m, 2H), 3.96 (t, J = 6.4 Hz, 2H), 3.02–2.87 (m, 3H),
2.82–2.73 (m, 2H), 2.24 (s, 3H), 1.88–1.80 (m, 2H), 1.06 (t,
J = 7.4 Hz, 3H); 13C NMR (300 MHz, CDCl3) d 178.82, 167.16,
155.77, 145.13, 140.85, 138.87, 135.21, 133.01, 130.82, 130.53,
130.45, 129.57, 129.38, 129.20, 128.92, 128.45, 127.75, 126.76,
126.44, 125.99, 125.88, 111.12, 69.51, 49.35, 40.31, 37.75, 36.94,
22.75, 20.40, 10.77; MS (FAB) 522 (M+H)+; calcd for C34H36NO4
calcd for C33H33FNO4 526.2394, found 526.2399; [
a
]
ꢁ9.4° (c
D
0.25, CH3CN). The enantiomeric excess (ee) of this compound
was determined to be 97% by chiral HPLC using a 4.6 ꢀ 250 mm
CHIRALPAK IA column (Diacel Chemical Industries, Ltd), eluting
at
a flow rate of 0.5 mL/min with 2-propanol/n-hexane/TFA
(1:2:0.03 v/v/v): retention time 19.8 min (the S enantiomer has a
retention time of 17.4 min); HPLC purity was estimated to be
99.9% by means of reversed-phase HPLC, using a Pegasil ODS
sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted on
a Shimadzu HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as the
eluant and detection at 254 nm.
522.2644, found 522.2660; [
a]
D ꢁ9.3° (c 0.25, CH3CN). The enantio-
meric excess (ee) of this compound was determined to be 97% by
chiral HPLC using a 4.6 ꢀ 250 mm CHIRALPAK IA column (Diacel
Chemical Industries, Ltd), eluting at a flow rate of 0.5 mL/min with
2-propanol/n-hexane/TFA (1:2:0.03 v/v/v): retention time
14.5 min (the S enantiomer has a retention time of 13.2 min); HPLC
purity was estimated to be 99.4% by means of reversed-phase
HPLC, using a Pegasil ODS sp100 column (4.6 ꢀ 250 mm, Senshu
Chemical, Japan) fitted on a Shimadzu HPLC system, with CH3CN/
0.1% TFA = 3:1 v/v as the eluant and detection at 254 nm.
4.6. 5-((R)-2-Benzyl-3-((S)-4-benzyl-2-oxooxazolidin-3-yl)-3-
oxopropyl)-2-propoxybenz-aldehyde oxime (8)
A mixture of 6 (3.31 g, 6.82 mmol), hydroxylamine hydrochlo-
ride (568 mg, 8.18 mmol), dehydrated pyridine (0.82 mL,
10.23 mmol) and 40 mL of ethanol was stirred overnight at room
temperature. Excess ethanol was evaporated, and the residue
was diluted with ethyl acetate, then washed with 1 mol/L HCl, a
saturated aqueous solution of sodium hydrocarbonate and brine.
The organic layer was separated, dried over anhydrous magnesium
sulfate and concentrated to afford 3.18 g of the title compound as a
colorless oil, which was used in the next step without further puri-
fication. 1H NMR (400 MHz, CDCl3) d 8.50 (s, 1H), 7.65 (d, J = 1.6 Hz,
1H), 7.28–7.15 (m, 9H), 7.00 (dd, J = 7.6, 2.0 Hz, 2H), 6.81 (d,
J = 8.4 Hz, 1H), 4.59–4.52 (m, 1H), 4.45–4.41 (m, 1H), 3.95–3.86
(m, 3H), 3.79 (t, J = 8.4 Hz, 1H), 3.08 (dd, J = 13.4. 8.2 Hz, 1H),
3.05–2.96 (m, 2H), 2.82 (dd, J = 13.4, 6.2 Hz, 1H), 2.78 (dd,
J = 13.6, 6.4 Hz, 1H), 2.49 (dd, J = 13.2, 9.2 Hz, 1H), 1.84–1.75 (m,
2H), 1.02 (t, J = 7.4 Hz, 3H); MS (FAB) 501 (M+H)+.
4.4. (R)-2-Benzyl-3-(3-((20-fluoro-[1,10-biphenyl]-4-
ylcarboxamido)methyl)-4-propoxy-phenyl)propanoic acid
((R)-7c)
This compound was prepared by means of a procedure similar
to that used for 7a.
1H NMR (400 MHz, CDCl3) d 7.80 (d, J = 8.4 Hz, 2H), 7.59 (dd,
J = 8.4, 1.6 Hz, 2H), 7.43 (dt, J = 7.6, 1.8 Hz, 1H), 7.37–7.32 (m,
1H), 7.28–7.14 (m, 8H), 6.80–6.76 (m, 2H), 4.65–4.56 (m, 2H),
3.95 (t, J = 6.6 Hz, 2H), 3.01–2.87 (m, 3H), 2.82–2.73 (m, 2H),
1.88–1.79 (m, 2H), 1.06 (t, J = 7.4 Hz, 3H); 13C NMR (300 MHz,
CDCl3) d 178.97, 166.96, 161.33, 158.04, 155.76, 138.86, 138.82,
138.81, 133.67, 130.82, 130.65, 130.61, 130.46, 129.67, 129.57,
129.21, 129.16, 129.12, 128.92, 128.45, 128.09, 127.91, 127.07,
126.43, 125.92, 124.52, 124.47, 116.37, 116.07, 111.12, 69.50,
49.37, 40.32, 37.77, 36.92, 22.75, 10.76; MS (FAB) 526 (M+H)+;
4.7. (S)-3-((R)-2-(3-(Aminomethyl)-4-propoxybenzyl)-3-
phenylpropanoyl)-4-benzyl-oxazolidin-2-one hydrochloride (9)
calcd for C33H33FNO4 526.2394, found 526.2399; [
a
]
ꢁ8.5° (c
A mixture of 8 (3.18 g, 6.35 mmol), 10% palladium on carbon
(1.60 g), concentrated HCl (8 mL) and 50 mL of ethanol was sub-
jected to catalytic hydrogenation at an initial hydrogen pressure
of 400–500 kPa. After completion of the reaction, the catalyst
was removed by filtration and the filtrate was concentrated to af-
ford 3.30 g of the title compound as a pale yellow solid, which was
used in the next step without further purification. MS (FAB) 487
(M+H)+.
D
0.25, CH3CN). The enantiomeric excess (ee) of this compound
was determined to be 98% by chiral HPLC using a 4.6 ꢀ 250 mm
CHIRALPAK IA column (Diacel Chemical Industries, Ltd), eluting
at
a flow rate of 0.5 mL/min with 2-propanol/n-hexane/TFA
(1:2:0.03 v/v/v): retention time 15.7 min (the S enantiomer has a
retention time of 14.6 min); HPLC purity was estimated to be
99.6% by means of reversed-phase HPLC, using a Pegasil ODS
sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted on
a Shimadzu HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as the
eluant and detection at 254 nm.
4.8. (R)-2-Benzyl-3-(4-propoxy-3-((4-(pyridin-2-
yl)benzamido)methyl)phenyl)propanoic acid ((R)-7g)
4.5. (R)-2-Benzyl-3-(3-((40-fluoro-[1,10-biphenyl]-4-ylcarboxamido)
methyl)-4-propoxy-phenyl)propanoic acid ((R)-7e)
Under an argon atmosphere, diethyl cyanophosphonate (DEPC)
(0.09 mL, 0.61 mmol) and triethylamine (0.18 mL, 1.30 mmol)
were added to a solution of compound 9 (246 mg, 0.47 mmol)
and 4-(2-pyridyl)benzoic acid (94 mg, 0.47 mmol) in dehydrated
N,N-dimethylformamide (10 mL) at 0 °C. The reaction mixture
This compound was prepared by means of a procedure similar
to that used for 7a.

M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
2329
was left to stand overnight, then poured into a saturated aqueous
solution of sodium hydrocarbonate, and the whole was extracted
with ethyl acetate. The extract was washed with brine, dried over
anhydrous magnesium sulfate and concentrated. The residue was
purified by silica gel column chromatography (eluant; n-hexane/
ethyl acetate = 1:2 v/v) to afford 170 mg (54%) of the intermediate
compound as a colorless oil. 1H NMR (400 MHz, CDCl3) d 8.71 (d,
J = 5.6 Hz, 1H), 8.00 (d, J = 8.4 Hz, 2H), 7.80–7.73 (m, 4H), 7.30–
7.15 (m, 11H), 6.90–6.87 (m, 2H), 6.82 (d, J = 8.4 Hz, 1H), 6.66 (t,
J = 5.4 Hz, 1H), 4.69 (dd, J = 14.6, 5.8 Hz, 1H), 4.64–4.55 (m, 2H),
4.41–4.35 (m, 1H), 4.00–3.92 (m, 2H), 3.90 (dd, J = 8.8, 2.4 Hz,
1H), 3.74 (t, J = 8.4 Hz, 1H), 3.08 (dd, J = 13.6, 8.8 Hz, 1H), 2.98
(dd, J = 13.2, 8.8 Hz, 1H), 2.92–2.79 (m, 3H), 2.51 (dd, J = 13.6,
8.8 Hz, 1H), 1.87–1.78 (m, 2H), 1.04 (t, J = 7.4 Hz, 3H); MS (FAB)
668 (M+H)+.
on a Shimadzu HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as the
eluant and detection at 254 nm.
4.10. (R)-2-Benzyl-3-(4-propoxy-3-((4-(thiophen-2-
yl)benzamido)methyl)phenyl)propanoic acid ((R)-7f)
This compound was prepared by means of a procedure similar
to that used for 7g and 7a.
1H NMR (400 MHz, DMSO-d6)
d 12.05 (s, 1H), 8.81 (t,
J = 5.6 Hz, 1H), 7.94 (t, J = 8.8 Hz, 2H), 7.76 (d, J = 8.4 Hz, 2H),
7.64 (dd, J = 3.6, 1.2 Hz, 1H), 7.62 (dd, J = 5.2, 1.2 Hz, 1H), 7.19–
7.15 (m, 3H), 7.12–7.09 (m, 3H), 7.02–6.99 (m, 2H), 6.86 (d,
J = 8.4 Hz, 1H), 4.49–4.38 (m, 2H), 3.92 (t, J = 6.2 Hz, 2H), 2.81–
2.71 (m, 3H), 2.68–2.56 (m, 2H), 1.77–1.68 (m, 2H), 0.98 (t,
J = 7.4 Hz, 3H); 13C NMR (300 MHz, DMSO-d6) d 175.64, 165.75,
154.51, 142.39, 139.42, 136.31, 133.18, 130.55, 128.79, 128.24,
128.23, 127.81, 126.83, 126.75, 126.13, 125.13, 124.92, 111.18,
69.06, 48.85, 37.74, 37.26, 36.79, 22.24, 10.64; MS (FAB) 514
This compound was treated with 30% hydrogen peroxide, as de-
scribed for 7a.
1H NMR (400 MHz, DMSO-d6) d 8.91 (t, J = 5.8 Hz, 1H), 8.74(d,
J = 4.4 Hz, 1H), 8.19–8.13 (m, 3H), 8.09–8.03 (m, 3H), 7.53 (t,
J = 6.0 Hz, 1H), 7.19–7.16 (m, 2H), 7.12–7.09 (m, 3H), 7.03–7.01
(m, 2H), 6.87 (d, J = 8.8 Hz, 1H), 4.51–4.40 (m, 2H), 3.93 (t,
J = 6.4 Hz, 2H), 2.80–2.70 (m, 3H), 2.67–2.57 (m, 2H), 1.77–1.69
(m, 2H), 0.98 (t, J = 7.2 Hz, 3H); 13C NMR (300 MHz, CDCl3) d
178.72, 166.91, 156.23, 155.71, 149.54, 141.71, 139.04, 137.22,
134.84, 131.02, 130.36, 129.21, 128.93, 128.40, 127.47, 127.11,
126.35, 125.79, 122.82, 121.19, 111.11, 69.48, 49.44, 40.32,
37.86, 36.94, 22.66, 10.75; MS (FAB) 509 (M+H)+; calcd for
(M+H)+; calcd for C31H32NO4 S 514.2051, found 514.2070; [
ꢁ8.6° (c 0.25, CH3CN). The enantiomeric excess (ee) of this com-
pound was determined to be 96% by chiral HPLC using
a]
D
a
4.6 ꢀ 250 mm CHIRALPAK IA column (Diacel Chemical Industries,
Ltd), eluting at a flow rate of 0.5 mL/min with 2-propanol/n-hex-
ane/TFA (1:2:0.03 v/v/v): retention time 17.6 min (the S enantio-
mer has
a retention time of 15.6 min); HPLC purity was
estimated to be 99.2% by means of reversed-phase HPLC, using
a Pegasil ODS sp100 column (4.6 ꢀ 250 mm, Senshu Chemical,
Japan) fitted on a Shimadzu HPLC system, with CH3CN/0.1%
TFA = 3:1 v/v as the eluant and detection at 285 nm.
C
32H33N2O4 509.2440, found 509.2421;
[
a
]
ꢁ7.3° (c 0.25,
D
CH3CN). The enantiomeric excess (ee) of this compound was
determined to be 97% by chiral HPLC using a 4.6 ꢀ 250 mm
CHIRALPAK IA column (Diacel Chemical Industries, Ltd), eluting
4.11. (R)-2-Benzyl-3-(4-propoxy-3-((4-(pyridin-3-yl)benzamido)
methyl)phenyl)propanoic acid ((R)-7h)
at
a flow rate of 0.5 mL/min with 2-propanol/n-hexane/TFA
(3:2:0.03 v/v/v): retention time 11.8 min (the S enantiomer has
a retention time of 10.8 min); HPLC purity was estimated to be
97.2% by means of reversed-phase HPLC, using a Pegasil ODS
sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted
on a Shimadzu HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as
the eluant and detection at 254 nm.
This compound was prepared by means of a procedure similar
to that used for 7g and 7a.
1H NMR (400 MHz, DMSO-d6) d 12.05 (s, 1H), 8.96 (d, J = 1.6 Hz,
1H), 8.88(t, J = 5.6 Hz, 1H), 8.60 (dd, J = 4.8, 1.6 Hz, 1H), 8.15 (dt,
J = 8.8, 1.6 Hz, 1H), 8.03 (d, J = 8.4 Hz, 2H), 7.86 (d, J = 8.4 Hz, 2H),
7.51 (dd, J = 8.0, 4.8 Hz, 1H), 7.19–7.09 (m, 5H), 7.03–7.00 (m,
2H), 6.87 (d, J = 8.0 Hz, 1H), 4.51–4.40 (m, 2H), 3.93 (t, J = 6.4 Hz,
2H), 2.80–2.70 (m, 3H), 2.67–2.57 (m, 2H), 1.78–1.68 (m, 2H),
0.99 (t, J = 7.4 Hz, 3H); 13C NMR (300 MHz, DMSO-d6) d 175.65,
165.84, 154.50, 149.60, 147.84, 139.69, 139.45, 134.38, 133.95,
130.58, 128.77, 128.21, 128.14, 128.04, 127.77, 126.85, 126.71,
126.11, 124.01, 111.18, 111.11, 69.06, 48.87, 37.76, 37.24, 36.80,
22.23, 10.63; MS (FAB) 509 (M+H)+; calcd for C32H33N2O4
4.9. (R)-2-Benzyl-3-(3-((30-fluoro-[1,10-biphenyl]-4-ylcarboxamido)
methyl)-4-propoxy-phenyl)propanoic acid ((R)-7d)
This compound was prepared by means of a procedure similar
to that used for 7g and 7a.
1H NMR (400 MHz, DMSO-d6) d 8.87 (t, J = 5.6 Hz, 1H), 8.00 (d,
J = 8.4, 2H), 7.82 (d, J = 8.8 Hz, 2H), 7.61–7.58 (m, 2H), 7.55–7.49
(m, 1H), 7.23 (dt, J = 8.8, 1.6 Hz, 1H), 7.19–7.15 (m, 2H), 7.12–
7.09 (m, 3H), 7.03–7.00 (m, 2H), 6.86 (d, J = 8.4 Hz, 1H), 4.50–
4.40 (m, 2H), 3.92 (t, J = 6.4 Hz, 2H), 2.79–2.71 (m, 3H), 2.69–2.56
(m, 2H), 1.77–1.69 (m, 2H), 0.99 (t, J = 7.4 Hz, 3H); 13C NMR
509.2440, found 509.2421; [
a]D ꢁ8.3° (c 0.25, CH3CN). The enantio-
meric excess (ee) of this compound was determined to be >98% by
chiral HPLC using a 4.6 ꢀ 250 mm CHIRALPAK IA column (Diacel
Chemical Industries, Ltd), eluting at a flow rate of 0.5 mL/min with
2-propanol/n-hexane/TFA (1:2:0.03 v/v/v): retention time 30.7 min
(the S enantiomer was not detectable); HPLC purity was estimated
to be 99.6% by means of reversed-phase HPLC, using a Pegasil ODS
sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted on a
Shimadzu HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as the elu-
ant and detection at 290 nm.
(300 MHz, CDCl3)
d 179.11, 166.86, 164.78, 161.52, 155.75,
142.74, 142.71, 142.29, 142.19, 138.85, 138.83, 130.84, 130.46,
130.45, 130.44, 130.34, 129.21, 128.90, 128.44, 127.57, 127.15,
126.43, 125.86, 122.83, 122.79, 114.88, 114.60, 114.21, 113.92,
111.13, 69.50, 49.40, 40.33, 37.77, 36.89, 22.74, 10.76; MS (FAB)
526 (M+H)+; calcd for C33H33FNO4 526.2394, found 526.2399;
[
a
]
ꢁ7.6° (c 0.25, CH3CN). The enantiomeric excess (ee) of this
D
4.12. (R)-2-Benzyl-3-(4-propoxy-3-((4-(pyridin-4-
yl)benzamido)methyl)phenyl)propanoic acid ((R)-7i)
compound was determined to be 96% by chiral HPLC using a
4.6 ꢀ 250 mm CHIRALPAK IA column (Diacel Chemical Industries,
Ltd), eluting at a flow rate of 0.5 mL/min with 2-propanol/n-hex-
ane/TFA (1:2:0.03 v/v/v): retention time 18.7 min (the S enantio-
mer has a retention time of 16.2 min). The purity was estimated
to be 99.7% by means of reversed-phase HPLC, using a Pegasil
ODS sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted
This compound was prepared by means of a procedure similar
to that used for 7g and 7a.
1H NMR (400 MHz, DMSO-d6) d 8.99 (t, J = 5.2 Hz, 1H), 8.88 (d,
J = 4.8 Hz, 2H), 8.23 (d, J = 5.2 Hz, 2H), 8.11–8.05 (m, 4H), 7.19–
7.09 (m, 5H), 7.03–7.00 (m, 2H), 6.87 (d, J = 8.0 Hz, 1H), 4.51–

2330
M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
4.41 (m, 2H), 3.93 (t, J = 6.2 Hz, 2H), 2.80–2.73 (m, 3H), 2.67–2.58
(m, 2H), 1.77–1.69 (m, 2H), 0.98 (t, J = 7.4 Hz, 3H); 13C NMR
(300 MHz, DMSO-d6) d 175.61, 165.51, 154.56, 139.40, 137.78,
136.25, 130.54, 128.78, 128.37, 128.24, 128.14, 127.88, 127.75,
126.54, 126.16, 111.24, 69.09, 48.82, 37.89, 37.26, 36.80, 22.24,
10.65; MS (FAB) 509 (M+H)+; calcd for C32H33N2O4 509.2440, found
98.1% by means of reversed-phase HPLC, using a Pegasil ODS
sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted
on a Shimadzu HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as
the eluant and detection at 254 nm.
4.15. (R)-2-Benzyl-3-(4-propoxy-3-((4-(pyrimidin-5-
yl)benzamido)methyl)phenyl)propanoic acid ((R)-7l)
509.2426; [
a]D ꢁ4.3° (c 0.25, DMSO). The enantiomeric excess (ee)
of this compound was determined to be >98% by chiral HPLC using
a 4.6 ꢀ 250 mm CHIRALPAK IA column (Diacel Chemical Industries,
Ltd), eluting at a flow rate of 0.5 mL/min with 2-propanol/n-hex-
ane/TFA (3:2:0.03 v/v/v): retention time 15.9 min (the S enantio-
mer was not visible); HPLC purity was estimated to be 99.6% by
means of reversed-phase HPLC, using a Pegasil ODS sp100 column
(4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted on a Shimadzu
HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as the eluant and
detection at 290 nm.
This compound was prepared by means of a procedure similar
to that used for 7g and 7a.
1H NMR (400 MHz, DMSO-d6) d 12.06 (s, 1H), 9.21 (s, 3H),
8.92 (t, J = 5.8 Hz, 1H), 8.06 (d, J = 8.8 Hz, 2H), 7.95 (d,
J = 8.4 Hz, 2H), 7.20–7.16 (m, 2H), 7.13–7.09 (m, 3H), 7.03–7.00
(m, 2H), 6.87 (d, J = 8.0 Hz, 1H), 4.51–4.41 (m, 2H), 3.93 (t,
J = 6.4 Hz, 2H), 2.80–2.71 (m, 3H), 2.67–2.57 (m, 2H), 1.78–1.69
(m, 2H), 0.99 (t, J = 7.2 Hz, 3H); 13C NMR (300 MHz, DMSO-d6)
d
175.61, 165.69, 157.71, 154.98, 154.52, 139.41, 136.41,
4.13. (R)-2-Benzyl-3-(4-propoxy-3-((4-(pyrimidin-2-
yl)benzamido)methyl)phenyl)propanoic acid ((R)-7j)
134.61, 132.44, 130.52, 128.76, 128.35, 128.22, 128.06, 127.82,
126.92, 126.64, 126.13, 111.19, 69.09, 48.85, 37.80, 37.25,
36.80, 22.22, 10.62; MS (FAB) 510 (M+H)+; calcd for
This compound was prepared by means of a procedure similar
to that used for 7g and 7a.
C
31H32N3O4 510.2393, found 510.2409;
[
a
]
ꢁ7.6° (c 0.25,
D
CH3CN). The enantiomeric excess (ee) of this compound was
determined to be 97% by chiral HPLC using a 4.6 ꢀ 250 mm
CHIRALPAK IA column (Diacel Chemical Industries, Ltd), eluting
1H NMR (400 MHz, DMSO-d6) d 12.06 (s, 1H), 8.94 (d, J = 4.8 Hz,
2H), 8.91 (t, J = 5.8 Hz, 1H), 8.47 (d, J = 8.8 Hz, 2H), 8.04 (d,
J = 8.4 Hz, 2H), 7.49 (t, J = 4.8 Hz, 1H), 7.19–7.16 (m, 2H), 7.12–
7.09 (m, 3H), 7.02–7.01 (m, 2H), 6.87 (d, J = 9.2 Hz, 1H), 4.50–
4.40 (m, 2H), 3.93 (t, J = 6.4 Hz, 2H), 2.82–2.71 (m, 3H), 2.70–2.58
(m, 2H), 1.77–1.69 (m, 2H), 0.98 (t, J = 7.4 Hz, 3H); 13C NMR
(300 MHz, DMSO-d6) d 175.64, 165.93, 162.63, 157.92, 154.52,
139.55, 139.43, 136.45, 130.59, 128.78, 128.21, 128.07, 127.89,
127.82, 127.57, 126.62, 126.11, 120.41, 111.18, 69.07, 48.86,
37.84, 37.28, 36.78, 22.22, 10.62; MS (FAB) 510 (M+H)+; calcd for
at
a flow rate of 0.5 mL/min with 2-propanol/n-hexane/TFA
(1:2:0.03 v/v/v): retention time 32.4 min (the S enantiomer has
a retention time of 29.9 min); HPLC purity was estimated to be
99.3% by means of reversed-phase HPLC, using a Pegasil ODS
sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Japan) fitted
on a Shimadzu HPLC system, with CH3CN/0.1% TFA = 3:1 v/v as
the eluant and detection at 254 nm.
C
31H32N3O4 510.2393, found 510.2409; [
a
]
ꢁ3.9° (c 0.25, CHCl3).
D
4.16. Thermodynamic aqueous solubility
The enantiomeric excess (ee) of this compound was determined
to be >98% by chiral HPLC using a 4.6 ꢀ 250 mm CHIRALPAK IA col-
umn (Diacel Chemical Industries, Ltd.), eluting at a flow rate of
0.5 mL/min with 2-propanol/n-hexane/TFA (1:2:0.03 v/v/v): reten-
tion time 17.5 min (the S enantiomer was not visible); HPLC purity
was estimated to be 98.4% by means of reversed-phase HPLC, using
a Pegasil ODS sp100 column (4.6 ꢀ 250 mm, Senshu Chemical, Ja-
Thermodynamic solubility determination was based on the
method of Avdeef and Testa. Briefly, about 1 mg of compound
was ground in an agate mortar and taken up in 1.0 mL of an equal
volume of a mixture of phosphate buffer (pH 7.2–7.4). The suspen-
sion was shaken for 48 h at 25 °C. An aliquot was filtered through a
Minisart RC 15 (0.45 lm). The filtrate was diluted in CH3CN and in-
pan) fitted on
TFA = 3:1 v/v as the eluant and detection at 254 nm.
a Shimadzu HPLC system, with CH3CN/0.1%
jected into an HPLC with UV detection at 254 nm. The concentra-
tion of the sample solution was calculated using a previously
determined calibration curve, corrected for the dilution factor of
the sample.
4.14. (R)-2-Benzyl-3-(4-propoxy-3-((4-(pyrazin-2-
yl)benzamido)methyl)phenyl)propanoic acid ((R)-7k)
4.17. Assays with GAL4-PPAR chimera receptors
This compound was prepared by means of a procedure similar
to that used for 7g and 7a.
1H NMR (400 MHz, DMSO-d6)
d 12.08 (s, 1H), 9.34 (d,
Receptor expression plasmid: An established chimeric receptor
system was utilized to allow comparison of the relative transcrip-
tional activity of the receptor subtypes. The mammalian
J = 1.2 Hz, 1H), 8.92 (t, J = 5.8 Hz, 1H), 8.75 (dd, J = 4.0, 1.6 Hz,
1H), 8.65 (d, J = 2.4 Hz, 1H), 8.25 (d, J = 8.8 Hz, 2H), 8.05 (d,
J = 8.4 Hz, 2H), 7.19–7.16 (m, 2H), 7.12–7.09 (m, 3H), 7.03–7.01
(m, 2H), 6.87 (d, J = 8.8 Hz, 1H), 4.51–4.41 (m, 2H), 3.93 (t,
J = 6.4 Hz, 2H), 2.80–2.71 (m, 3H), 2.69–2.57 (m, 2H), 1.77–1.69
(m, 2H), 0.98 (t, J = 7.4 Hz, 3H); 13C NMR (300 MHz, CDCl3) d
178.61, 166.61, 155.78, 151.75, 144.36, 143.34, 142.18, 138.91,
138.86, 135.81, 130.97, 130.49, 129.28, 128.92, 128.46, 127.70,
127.04, 126.44, 125.79, 111.19, 69.55, 49.37, 40.42, 37.87,
36.95, 22.76, 10.77; MS (FAB) 510 (M+H)+; calcd for
expression vectors pSG5-GAL4-hPPAR
pSG5-GAL4-hPPARd, which express the ligand binding domains
(LBDs) of human PPAR (amino acids 167–468), PPAR 1 (amino
a, pSG5-GAL4-hPPARc and
a
c
acids 176–477) and PPARd (amino acids 139–441) each fused to
the yeast transcription factor GAL4 DNA binding domain (amino
acid 1–147), were used. Reporter plasmid: MH100x4-tk-luc45
was used as the reporter plasmid.
4.18. Transient transfection assays
C31H32N3O4 510.2393, found 510.2409;
[
a
]
ꢁ8.3° (c 0.25,
D
CH3CN). The enantiomeric excess (ee) of this compound was
determined to be >98% by chiral HPLC using a 4.6 ꢀ 250 mm
CHIRALPAK IA column (Diacel Chemical Industries, Ltd), eluting
The African green monkey kidney cell line CV-1 was used for
the transfection assay. CV-1 cells were seeded in 24-well plates
at 0.5 ꢀ 105 cells per well and cultured for 24 h. Transfection mix-
tures for chimeric receptors contained 30 ng of receptor expression
plasmid, 250 ng of the reporter plasmid, 350 ng of pCMX-b-galac-
at
a flow rate of 0.5 mL/min with 2-propanol/n-hexane/TFA
(1:2:0.03 v/v/v): retention time 25.6 min (the S enantiomer has
a retention time of 20.4 min); HPLC purity was estimated to be

M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
2331
tosidase (bGAL) expression plasmid as a control for transfection
efficiency, 120 ng of pGEM4 carrier plasmid and 2 L of a lipofec-
previously reported structures as probes. The correctly positioned
molecules were refined with CNS and O. The initial atomic model
of (R)-7j was built using MOE (Ryoka Systems Inc.) and topology
and parameter files for the refinement were generated by the
HIC-Up server.
l
tion reagent (Lipofectamine 2000, Invitrogen). The mixtures were
added to cells and incubated for 5 h according to the manufac-
turer’s instructions. After the transfection, cells were incubated
for an additional 40 h in the presence of new compounds or refer-
ence compounds. Cell lysates were prepared with a lysis buffer
(Passive Lysis Buffer, Promega) and used in the luciferase and bGAL
assays. The luciferase and b-GAL activity were measured according
to the methods of Umesono et al. with slight modifications. A sub-
strate reagent kit (Picagene, Toyo Ink) was used for the luciferase
assay.
The structure of hPPAR
replacement method using Molrep. Apo PPAR
c
LBD was solved by the molecular
LBD structure
c
(PDB code 1PRG) was used as a search model and refined with Ref-
mac5. The crystallographic data and data collection statistics of all
the crystals are provided in Table 1.
Acknowledgements
4.19. Calculation of relative PPAR transactivation activities
This work was supported in part by the Targeted Proteins Re-
search Program of the Japan Science and Technology Corporation
(JST), the Uehara Memorial Foundation, the Tokyo Biochemical Re-
search Foundation (TBRF), and the Okayama Foundation for Sci-
ence and Technology (OFST). The authors would like to thank
Nippon Chemiphar Co. Ltd. for transient transfection assays.
Relative PPAR transcriptional activity with respect to maximal
activity was calculated based on the luciferase activity of cells trea-
ted with a positive control [10ꢁ6 M GW-590735 (PPAR
a
selective
, 10ꢁ5 M rosiglitazone maleate (PPAR
selec-
tive agonist) for hPPAR
and 10ꢁ7 M GW-501516 (PPARd selective
agonist) for hPPAR
a
c
c
agonist) for hPPARd], as the maximal activity, and the luciferase
activity of cells treated with 0.1% DMSO, taken as the minimum
activity.
References and notes
1. Lemberger, T.; Desvergne, B.; Wahli, W. Annu. Rev. Cell Dev. Biol. 1996, 12, 335.
2. Kubota, T.; Koshizuka, K.; Asou, H.; Williamson, E. A.; Said, J. W.; Holded, S.;
Miyoshi, I.; Koeffler, H. P. Cancer Res. 1998, 58, 3344.
4.20. Calculation of EC50 values
3. Kitamura, S.; Miyazaki, Y.; Shinomura, Y.; Kondo, S.; Kanayama, S.; Mastuzawa,
Y. Jpn. J. Cancer Res. 1999, 90, 75.
4. Chawla, A.; Schwarz, E. J.; Dimaculangan, D. D.; Lazar, M. A. Endocrinology 1994,
135, 798.
5. Tontonoz, P.; Nagy, L.; Alvarez, J. G. A.; Thomazy, V. A.; Evans, R. M. Cell 1998,
93, 241.
6. Hu, E.; Tontonoz, P.; Spiegelman, B. M. Proc. Natl. Acad. Sci. USA 1995, 92, 9856.
7. Mendez, M.; LaPointe, M. C. Hypertension 2003, 42, 844.
8. Wang, N.; Yin, R.; Liu, Y.; Mao, G.; Xi, F. Circ. J. 2011, 75, 528.
9. Kaundal, R.; Sharma, S. S. Drug News Perspect. 2010, 23, 241.
10. Landreth, G. Curr. Alzheimer Res. 2007, 4, 159.
11. Morrison, R. F.; Farmer, S. R. J. Biol. Chem. 1999, 274, 17088.
12. Chopita, N.; Andoni, N.; Ross, A.; Villaverde, A. Endosc. Clin. N. Am. 2007, 17,
533.
EC50 values, defined as the concentration of test compounds and
positive controls (GW-590735, rosiglitazone maleate and GW-
501516 required to produce 50% of maximal reporter activity, were
calculated with Prism software (Graphpad Software).
For comparison, GW-590735 activated human GAL4 PPAR
with an EC50 = 0.010 M, rosiglitazone activated human GAL4
PPAR with an EC50 = 0.1 M and GW-501516 activated human
a
l
c
l
GAL4 PPARd with an EC50 = 0.0013 lM.
13. Ikeguchi, M.; Miyake, T.; Matsunaga, T.; Yamamoto, M.; Fukumoto, Y.; Yamada,
Y.; Fukuda, K.; Saito, H.; Tatebe, S.; Tsujitani, S. Surg. Today 2009, 39, 290.
14. Ban, S.; Kasuga, J.; Nakagome, I.; Nobusada, H.; Takayama, F.; Hirono, S.;
Kawasaki, H.; Hashimoto, Y.; Miyachi, H. Bioorg. Med. Chem. 2011, 19, 3183.
15. Nomura, M.; Tanase, T.; Ide, T.; Tsunoda, M.; Suzuki, M.; Uchiki, U.; Murakami,
K.; Miyachi, H. J. Med. Chem. 2003, 14, 581.
16. Kasuga, J.; Yamasaki, D.; Araya, Y.; Nakagawa, A.; Makishima, M.; Doi, T.;
Hashimoto, Y.; Miyachi, H. Bioorg. Med. Chem. 2006, 14, 8405.
17. Araya, Y.; Kasuga, J.; Toyota, K.; Hirakawa, Y.; Oyama, T.; Makishima, M.;
Morikawa, K.; Hashimoto, Y.; Miyachi, H. Chem. Pharm. Bull. 2008, 56, 1357.
18. Kasuga, J.; Nakagome, I.; Aoyama, A.; Sako, K.; Ishizawa, M.; Ogura, M.;
Makishima, M.; Hirono, S.; Hashimoto, Y.; Miyachi, H. Bioorg. Med. Chem. 2007,
15, 5177.
19. Kasuga, J.; Yamasaki, D.; Ogura, K.; Shimizu, M.; Sato, M.; Makishima, M.; Doi,
T.; Hashimoto, Y.; Miyachi, H. Bioorg. Med. Chem. Lett. 2008, 18, 1110.
20. Ohashi, M.; Oyama, T.; Nakagome, I.; Sato, M.; Nishio, Y.; Nobusada, H.; Hirono,
S.; Morikawa, K.; Hashimoto, Y.; Miyachi, H. J. Med. Chem. 2011, 54, 331.
21. Dube, D.; Scholte, A. Tetrahedron Lett. 1999, 40, 22958.
22. Avdeef, A.; Testa, B. Cell. Mol. Life Sci. 2002, 59, 1681.
4.21. Adipogenesis assay
Mouse adipogenic fibroblast cells, 3T3-L1, were cultured in
DMEM with 10% FBS. After the cells had become confluent, they
were pretreated with 1
tyl-1-methylxanthine (IBMX), and 5
l
M dexamethasone (Dex), 0.5 mM 3-isobu-
g/mL insulin to initiate adi-
l
pogenesis for 2 days. Various ligands were added to the cells.
After 6 days, the cells were stained with Oil Red O, and the degree
of adipogenesis was quantitatively measured at OD550. The error
bars represent the standard deviations (SD).
4.22. Protein expression, purification, and crystallization
The recombinant ligand-binding domain (LBD) of human PPAR
c
(residues 203–477) was expressed as an N-terminal His-tagged
protein in E. coli BL21 (DE3), using the pET28a vector (Novagen).
The protein was purified by means of several steps of column chro-
matography. Apo PPARc LBD crystals were soaked in ligand solu-
tion for two or three weeks.
23. Umesono, K.; Murakami, K.; Thompson, C. C.; Evans, R. M. Cell 1991, 65, 1255.
24. Oyama, T.; Toyota, K.; Waku, T.; Hirakawa, Y.; Nagasawa, N.; Kasuga, J.;
Hashimoto, Y.; Miyachi, H.; Morikawa, K. Acta Crystallogr. D Biol. Crystallogr.
2009, 65, 786.
25. Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
26. Brünger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.; Gros, P.; Grosse-
Kunstleve, R. W.; Jiang, J. S.; Kuszewski, J.; Nilges, M.; Pannu, N. S.; Read, R. J.;
Rice, L. M.; Simonson, T.; Warren, G. L. Acta Cryst. 1998, D54, 905.
27. Jones, T. A.; Zou, J. Y.; Cowan, S. W.; Kjeldgaard, M. Acta Cryst. 1991, A47, 110.
28. Kleywegt, G. J. Acta. Cryst. D 2007, 63, 940.
4.23. Diffraction data collection
29. Waku, T.; Shiraki, T.; Oyama, T.; Maebara, K.; Nakamori, R.; Morikawa, K. EMBO
J. 2010, 29, 3395.
30. Yashiro, M.; Chung, Y. S.; Nishimura, S.; Inoue, T.; Sowa, M. Clin. Exp. Metastasis
1996, 14, 43.
31. Bai, L.; Mihara, K.; Kondo, Y.; Honma, M.; Namba, M. Int. J. Cancer 1993, 53, 451.
32. Than, S. S.; Kataoka, K.; Sakaguchi, M.; Murata, H.; Abarzua, F.; Taketa, C.; Du,
G.; Yashiro, M.; Yanagihara, K.; Nasu, Y.; Kumon, H.; Huh, N. H. Oncol. Rep.
2011, 25, 989.
The diffraction data for hPPAR
a
LBD complexed with (R)-7j
were collected on BL38B1 at SPring-8 (Harima, Japan) and were
processed using HKL-2000. Diffraction data were processed with
HKL2000.
4.24. Structure analysis and refinement
33. Kasuga, J.; Ishikawa, M.; Yonehara, M.; Makishima, M.; Hashimoto, Y.; Miyachi,
H. Bioorg. Med. Chem. 2010, 18, 7164.
34. ClogP values were calculated with Chemdraw Ultra ver. 10.0.
The structure of hPPAR
the molecular-replacement method with the program CNS using
a LBD complex with (R)-7j was solved by

2332
M. Ohashi et al. / Bioorg. Med. Chem. 21 (2013) 2319–2332
35. Tontonoz, T.; Hu, E.; Spiegelman, B. M. Cell 1994, 79, 1147.
36. Kuwabara, N.; Oyama, T.; Tomioka, D.; Ohashi, M.; Yanagisawa, J.; Shimizu, T.;
Miyachi, H. J. Med. Chem. 2012, 55, 893.
44. Montanari, R.; Saccoccia, F.; Scotti, E.; Crestani, M.; Godio, C.; Gilardi, F.;
Loiodice, F.; Fracchiolla, G.; Laghezza, A.; Tortorella, P.; Lavecchia, A.; Novellino,
E.; Mazza, F.; Aschi, M.; Pochetti, G. J. Med. Chem. 2008, 51, 7768.
45. Oberfield, J. L.; Collins, J. L.; Holmes, C. P.; Goreham, D. M.; Cooper, J. P.; Cobb, J.
E.; Lenhard, J. M.; Hull-Ryde, E. A.; Mohr, C. P.; Blanchard, S. G.; Parks, D. J.;
Moore, L. B.; Lehmann, J. M.; Plunket, K.; Miller, A. B.; Milburn, M. V.; Kliewer, S.
A.; Willson, T. M. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 6102.
37. Breslow, E.; Gargiulo, P. Biochemistry 1977, 16, 3397.
38. Leo, C.; Yang, X.; Liu, J.; Li, H.; Chen, J. D. J. Biol. Chem. 2001, 276, 23127.
39. Benko, S.; Love, J. D.; Beládi, M.; Schwabe, J. W.; Nagy, L. J. Biol. Chem. 2003, 278,
43797.
40. Henke, B. R.; Blanchard, S. G.; Brackeen, M. F.; Brown, K. K.; Cobb, J. E.; Collins, J.
L., ; Harrington, W. W., Jr.; Hashim, M. A.; Hull-Ryde, E. A.; Kaldor, I.; Kliewer, S.
A.; Lake, D. H.; Leesnitzer, L. M.; Lehmann, J. M.; Lenhard, J. M.; Orband-Miller,
L. A.; Miller, J. F.; Mook, R. A., Jr.; Noble, S. A.; Oliver, W., Jr.; Parks, D. J.; Plunket,
K. D.; Szewczyk, J. R.; Willson, T. M. J. Med. Chem. 1998, 41, 5020.
41. Gampe, R. T., Jr.; Montana, V. G.; Lambert, M. H.; Miller, A. B.; Bledsoe, R.
K.; Milburn, M. V.; Kliewer, S. A.; Willson, T. M.; Xu, H. E. Mol. Cell 2000,
5, 545.
46. Lu, I. L.; Huang, C. F.; Peng, Y. H.; Lin, Y. T.; Hsieh, H. P.; Chen, C. T.; Lien, T. W.;
Lee, H. J.; Mahindroo, N.; Prakash, E.; Yueh, A.; Chen, H. Y.; Goparaju, C. M.;
Chen, X.; Liao, C. C.; Chao, Y. S.; Hsu, J. T.; Wu, S. Y. J. Med. Chem. 2006, 49, 2703.
47. Amato, A. A.; Rajagopalan, S.; Lin, J. Z.; Carvalho, B. M.; Figueira, A. C.; Lu, J.;
Ayers, S. D.; Mottin, M.; Silveira, R. L.; Souza, P. C.; Mourão, R. H.; Saad, M. J.;
Togashi, M.; Simeoni, L. A.; Abdalla, D. S.; Skaf, M. S.; Polikparpov, I.; Lima, M.
C.; Galdino, S. L.; Brennan, R. G.; Baxter, J. D.; Pitta, I. R.; Webb, P.; Phillips, K. J.;
Neves, F. A. J. Biol. Chem. 2012, 287, 28169.
42. Root Mean Square Deviations were calculated with Molecular Operating
Environment (MOE).
48. Takahashi, N.; Okumura, T.; Motomura, W.; Fujimoto, Y.; Kawabata, I.; Kohgo,
Y. FEBS Lett. 1999, 455, 135.
43. Pochetti, G.; Godio, C.; Mitro, N.; Caruso, D.; Galmozzi, A.; Scurati, S.; Loiodice,
F.; Fracchiolla, G.; Tortorella, P.; Laghezza, A.; Lavecchia, A.; Novellino, E.;
Mazza, F.; Crestani, M. J. Biol. Chem. 2007, 282, 17314.
49. Cheon, C. W.; Kim, D. H.; Cho, Y. H.; Kim, J. H. World J. Gastroenterol. 2009, 15,
310.