Discovery of a novel class of 1,3-dioxane-2-carboxylic acid derivatives as subtype-selective peroxisome proliferator-activated receptor α (PPARα) agonists

Article abstract of DOI:10.1016/j.bmcl.2008.01.086

A new series of 1,3-dioxane-2-carboxylic acid derivatives was synthesized and evaluated for agonist activity at human peroxisome proliferator-activated receptor (PPAR) subtypes. Structure-activity relationship studies led to the identification of 2-methyl-c-5-[4-(5-methyl-2-phenyl-1,3-oxazol-4-yl)butyl]-1,3-dioxane-r-2-carboxylic acid 4b as a potent PPARα agonist with high subtype selectivity at human receptor subtypes. This compound exhibited a substantial hypolipidemic effect in type 2 diabetic KK-A^y mice.

Full text of DOI:10.1016/j.bmcl.2008.01.086

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2128–2132
Discovery of a novel class of 1,3-dioxane-2-carboxylic
acid derivatives as subtype-selective peroxisome
proliferator-activated receptor a (PPARa) agonists
Tomiyoshi Aoki, Tetsuo Asaki,* Taisuke Hamamoto, Yukiteru Sugiyama, Shinji Ohmachi,
Kenji Kuwabara, Kohji Murakami and Makoto Todo
Discovery Research Laboratories, Nippon Shinyaku Co., Ltd, 14 Nishinosho-Monguchi-Cho,
Kisshoin, Minami-Ku, Kyoto 601-8550, Japan
Received 3 August 2007; revised 11 January 2008; accepted 23 January 2008
Available online 30 January 2008
Abstract—A new series of 1,3-dioxane-2-carboxylic acid derivatives was synthesized and evaluated for agonist activity at human
peroxisome proliferator-activated receptor (PPAR) subtypes. Structure-activity relationship studies led to the identification of 2-
methyl-c-5-[4-(5-methyl-2-phenyl-1,3-oxazol-4-yl)butyl]-1,3-dioxane-r-2-carboxylic acid 4b as a potent PPARa agonist with high
subtype selectivity at human receptor subtypes. This compound exhibited a substantial hypolipidemic effect in type 2 diabetic
KK-Ay mice.
Ó 2008 Elsevier Ltd. All rights reserved.
Peroxisome proliferator-activated receptors (PPARs)
are ligand-activated transcription factors and members
of the thyroid/steroid hormone nuclear receptor super-
family.1 Three subtypes, PPARa, PPARc and PPARd,
exhibit different tissue distributions and play important
roles in lipid, lipoprotein and glucose homeostasis.
PPARa, abundant in metabolically active tissues such
as liver, heart, kidney and muscle, regulates fatty acid
catabolism.2 Hypolipidemic fibrate drugs (fibrates), an
important class of PPARa ligands, lower serum triglyc-
eride levels and moderately raise high-density lipopro-
tein (HDL) cholesterol levels in patients with
hyperlipidemia. However, recent molecular-pharmaco-
logical studies demonstrate that fibrates, such as fenofi-
brate 1a and bezafibrate 2 (Fig. 1), show significant
cross-reactivity with other PPAR subtypes.3 For exam-
ple, fenofibric acid 1b, the active metabolite of fenofi-
brate 1a, is a dual agonist for PPARa and PPARc,
while bezafibrate 2 activates all three PPAR subtypes
at comparable doses. In addition, the agonist activities
of these two compounds at PPARa are weak, and the
clinical doses required to treat hyperlipidemia are rela-
O
CO2H
Cl
O
CO2R
Cl
O
N
H
O
1a: R = i-Pr, fenofibrate
1b: R = H, fenofibric acid
2: bezafibrate
Figure 1. Chemical structures of fibrates.
tively high (200–400 mg/day in Japan). Therefore, more
potent and selective agonists for PPARa are expected to
have superior therapeutic efficacy in treating these con-
ditions.4 Although some potent and selective PPARa
agonists have already been reported,5–8 they are not
yet in clinical use.
During our efforts to identify compounds with lipid-low-
ering effects in KK-Ay mice, we found 2-methyl-2-[6-(5-
methyl-2-phenyl-4-oxazolyl)hexyloxy]propionic acid 3
(Fig. 2) to have potent lipid-lowering activity. (KK-Ay
mice display severe hyperglycemia, hyperinsulinemia,
hypertriglyceridemia and obesity, the typical symptoms
of type 2 diabetes, and are widely used as animal models
of type 2 diabetes.9) When orally administered at a dose
of 1 mg/kg once a day for 4 days, this compound low-
ered plasma triglyceride levels by 32%. The compound
has a flexible conformation due to the long alkylene
Keywords: PPARa agonist; 1,3-Dioxane carboxylic acid; Metabolic
disorder.
*
Corresponding author. Tel.: +81 75 321 9167; fax: +81 75 321
9039; e-mail: t.asaki@po.nippon-shinyaku.co.jp
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.01.086

T. Aoki et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2128–2132
2129
O
N
O
N
O
O
CO2H
cyclize
CO2H
O
3
4b
Figure 2. Structural modification of 3.
chain linking the oxazole ring and carboxyl group, and
it seemed to us that the potency of the compound could
be enhanced by restricting its conformation. To test this
hypothesis, we designed and synthesized 1,3-dioxane-2-
carboxylic acid derivative 4b (Fig. 2), whose conforma-
tion is significantly restricted by cyclization of the
carboxylic acid tail of 3. As expected, this compound
exhibited superior in vivo pharmacological activity.
During our investigation of their lipid-lowering activity,
we found that some 1,3-dioxane-2-carboxylic acid deriv-
atives had strong and highly subtype-selective agonist
activity at human PPARa. To gain a better understand-
ing of their pharmacological characteristics, we investi-
gated the effect of the structure of the linker and of
the dioxane moiety on their PPAR subtype agonist
activities.
7a–c. Cyclodehydration12 of 7a–c by treatment with
phosphorus oxychloride gave the corresponding oxaz-
oles, which after reduction with lithium aluminium
hydride followed by bromination of the resulting alco-
hols gave the bromides 8a–c. Alkylation of diethyl
malonate with 8a–c yielded the diesters, which were con-
verted to the dialcohols 9a–c by reduction with lithium
aluminium hydride. Formation of the dioxane ring from
the diols 9a–c with methyl pyruvate on refluxing with
boron trifluoride in acetonitrile yielded a mixture of geo-
metrical isomers, which was resolved by column chro-
matography to afford the cis-isomers 10a–c and one
trans-isomer 11b. The geometrical isomers 10b and 11b
have different chemical-shift patterns of the methylene
protons at the 4- and 6-positions and the methine proton
at the 5-position of the 1,3-dioxane ring moiety, which
are identical to those observed by Harabe et al.13 The
configurations of the protons were also confirmed by
NOE measurements.14 The ratio of cis to trans isomers
was approximately 3:1 by gas chromatographic analysis.
Finally, the target compounds 4a–c and 12b were ob-
tained by alkaline hydrolysis of the esters 10a–c and
11b. Compounds 13–15, which have different substitu-
ents at the 2-position of the 1,3-dioxane ring, were pre-
pared from 9b and the requisite a-ketoesters as outlined
in Scheme 2. The configuration of the compounds was
determined to be cis from the chemical-shift patterns
The general method for the synthesis of compounds 4a–
c and 12b with an alkylene tether between the oxazole
ring and the dioxane ring is shown in Scheme 1. Diethyl
2-(benzoylamino)malonate 510 was alkylated with one of
three commercially available ethyl bromoalkylcarboxy-
lates, and subsequent alkaline hydrolysis and decar-
boxylation gave the dicarboxylic acids 6a–c. After
Dakin–West conversion11 of compounds 6a–c to the
corresponding methyl ketones, esterification of the
remaining carboxyl group afforded the methyl esters
O
O
CO2Et
CO2Et
O
CO2H
(CH2)nCO2H
O
a
c
b
O
N
H
N
N
(CH2)nCO2Me
H
H
N
(CH2)nCH2Br
5
6a: n = 2
6b: n = 3
6c: n = 4
7a: n = 2
7b: n = 3
7c: n = 4
8a: n = 2
8b: n = 3
8c: n = 4
O
O
N
O
N
d
e
f
N
(CH2)nCH2
(CH2)nCH2
(CH2)nCH2
O
O
O
OH
CO2Me
CO2H
O
9a: n = 2
9b: n = 3
9c: n = 4
10a: n = 2
10b: n = 3
10c: n = 4
4a: n = 2
4b: n = 3
4c: n = 4
OH
Me
Me
11b (trans-isomer of 10b)
12b (trans-isomer of 4b)
Scheme 1. Reagents: (a) i—Br(CH2)nCO2Et, NaOEt, EtOH, ii—aq NaOH, EtOH, iii—AcOEt, xylene; (b) i—Ac2O, pyridine, ii—H2O, iii—H2SO4,
MeOH; (c) i—POCl3, toluene, ii—LiAlH4, Et2O, iii—CBr4, Et2O; (d) i—CH2(CO2Et)2, NaH, THF, ii—LiAlH4, Et2O; (e) i—MeCOCO2Me,
BF3ÆEt2O, MeCN; (f) aq NaOH, MeOH.
O
N
a, b
O
N
OH
OH
O
CO2H
9b
O
13: R = H
14: R = Et
15: R = -Bu
R
i
Scheme 2. Reagents: (a) i—RCOCO2Me, BF3ÆEt2O, MeCN, (b) aq NaOH, MeOH.

2130
T. Aoki et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2128–2132
of the 1,3-dioxane ring moiety of the corresponding
esters or carboxylic acids, judging from their similarity
to the H NMR spectrum of 10b and the observations
compound, 4a, which was inactive at all PPAR sub-
types. In contrast, increasing the length of the linker
gave a compound, 4c, whose PPARa agonist activity
was enhanced threefold compared to 4b. Compound
4c was found to be an essentially equipotent dual
PPARa/PPARc agonist with activation characteristics
more similar to 3 than to 4b, indicating that the length
of the linker in this series has a dramatic effect on
PPARc agonist activity. We also synthesized four com-
pounds 21a–d in which a phenyl ring is inserted into the
side-chain tether of 4b to restrict its conformational flex-
ibility. Although 21a, with its meta-substituted phenyl
linker, showed PPARa selectivity, its PPARa agonist
activity was 10-fold lower than that of 4b. On the other
hand, para-substituted variant 21b had substantially less
activity at all PPAR subtypes. Increasing the length of
the linker in 21a,b by one methylene unit led to 21c,d,
which moderately activated both the a and the c sub-
types. These results strongly support the notion that
the potency and selectivity for PPARa are highly depen-
dent on the spatial relationship between the phenyloxaz-
ole moiety and the carboxyl group.
1
of Harabe et al.13
The synthesis of compounds 21a–d, in which a phenyl
ring is included within the methylene linker, is shown
in Scheme 3. The starting materials 16a,b and 16c,d were
prepared from ethyl 3- or 4-bromophenylacetate and 3-
or 4-bromobenzylbromide, respectively, according to a
known method.15,16 Compounds 16a–d were reduced
with diisobutylaluminium hydride and then protected
by acetonide formation to give the dioxanes 17a–d.
The oxazole 18 was easily derived from 4-chloro-
methyl-5-methyl-2-phenyloxazole17 by refluxing with so-
dium iodide in acetone. The coupling of 17a–d and 18
was accomplished by a procedure similar to that previ-
ously outlined18,19 to give 19a–d. After deprotection by
cleavage of the acetonides 19a–d with pyridinium p-tol-
uenesulfonate, the final compounds 21a–d were pre-
pared through 20a–d in a stepwise manner similar to
that described for the preparation of 4a–c.
The newly synthesized compounds were screened for
activity against each of the human PPAR subtypes by
using an established cell-based transactivation assay in
CV-1 fibroblast cells as described previously.20 The ini-
tial lead compound 3 was found to be a potent dual ago-
nist of PPARa (EC50 = 0.3 lM) and PPARc
EC50 = 0.3 lM) with 100-fold selectivity over PPARd
(EC50 = 30 lM). Unexpectedly, the dioxane 4b derived
from 3 had even higher agonist activity at PPARa,
whereas it activated PPARc and PPARd hardly at all.
Compound 4b is a potent and selective agonist of human
PPARa with more than 300-fold selectivity over the c
and d subtypes. The trans-isomer 12b was 10-fold less
potent as an agonist of PPARa, a characteristic which
reduces its PPARa selectivity.
To confirm the requirement for a substituent at the 2-
position of the 1,3-dioxane ring, we next explored a lim-
ited number of compounds 13–15 (Table 2). Com-
pounds 14 (R = Et) and l5 (R = i-Bu) retained potent
PPARa activity and selectivity, with 10 times the po-
tency of 4b and equal PPARa selectivity, so that med-
ium-sized alkyl groups were well tolerated. In contrast,
replacement of the 2-position methyl group with hydro-
gen afforded a compound, 13, that was inactive at all
three PPAR subtypes. These data indicate that an alkyl
group with suitable bulk at this position is indispensable
for the display of potent PPARa agonist activity.
We examined in type 2 diabetic KK-Ay mice the phar-
macological effects of two compounds, 4b and 14, with
excellent human PPARa agonist activity and subtype
selectivity (Table 3). Even though they activated mouse
PPARa threefold less potently than they did human
PPARa, they had sufficient in vitro activity (4b,
EC50 = 1 lM; 14, EC50 = 0.1 lM) with high selectivity
We next turned our attention to the conformationally
flexible alkylene linker of 4b and investigated the effects
of its length and rigidity (Table 1). Reducing the length
of the alkylene linker of 4b by one methylene unit gave a
R1
R2
R1
R2
O
O
N
4
3
O
CO2Et
CO2Et
b
a
O
(CH2)n
(CH2)n
O
O
N
(CH2)n
2
I
16a: n = 0, R1 = H, R2 = Br
16b: n = 0, R1 = Br, R2 = H
16c: n = 1, R1 = H, R2 = Br
16d: n = 1, R1 = Br, R2 = H
17a: n = 0, R1 = H, R2 = Br
17b: n = 0, R1 = Br, R2 = H
17c: n = 1, R1 = H, R2 = Br
17d: n = 1, R1 = Br, R2 = H
19a: 3-position, n = 0
19b: 4-position, n = 0
19c: 3-position, n = 1
19d: 4-position, n = 1
18
CO2Me
CO2H
O
O
4
3
O
O
N
4
3
c
d
Me
O
Me
N
O
(CH2)n
(CH2)n
2
2
20a: 3-position, n = 0
20b: 4-position, n = 0
20c: 3-position, n = 1
20d: 4-position, n = 1
21a: 3-position, n = 0
21b: 4-position, n = 0
21c: 3-position, n = 1
21d: 4-position, n = 1
Scheme 3. Reagents: (a) i—diisobutylaluminium hydride, toluene, ii—acetone, TsOH, benzene; (b) n-BuLi, n-Bu3PÆCuI, N,N,N0,N0-tetramethyl-
ethylenediamine, THF; (c) i—pyridinium p-toluenesulfonate, EtOH, ii—MeCOCO2Me, BF3ÆEt2O, MeCN; (d) aq NaOH, MeOH.

T. Aoki et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2128–2132
2131
Table 1. Effects of linker variations in cis-1,3-dioxane-2-carboxylic acid derivatives on human PPAR agonist activity
O
Linker
N
O
CO2H
O
Compound
Linker
Human PPAR subtype (EC50, lM)a,b
a
c
d
3
(Lead compound)
0.3
0.3
30
4a
4b
4c
–(CH2)3–
–(CH2)4–
–(CH2)5–
(8%)c
0.3
(2%)
100
0.3
(12%)
(0%)
(34%)
0.1
21a
3
(12%)
(7%)
—CH2
—CH2
21b
21c
21d
(5%)
(10%)
(0%)
3
3
1
3
(28%)
(38%)
—CH2
—CH2
CH2—
—
CH2
12b
1b
2
–(CH2)4– (trans-1,3-dioxane)
(Fenofibric acid)
(Bezafibrate)
3
30
>100
100
(10%)
>100
30
30
30
a The agonist activity of each PPAR subtype was measured by the corresponding transactivation assay in transiently transfected CV-1 cells (n = 2).
b EC50, the concentration of test compounds that gave 50% of the maximum transactivation activity of each positive control.
c Less than 50% of maximum activity at 10 lM. The percentage of maximum activation observed at 10 lM is shown in parentheses.
Table 2. Effects of substituents at the 2-position of the 1,3-dioxane
ring in 1,3-dioxane-2-carboxylic acid derivatives on human PPAR
agonist activity
Table 3. Hypolipidemic and hypoglycemic effects of 1,3-dioxane-2-
carboxylic acid derivatives 4b and 14 in diabetic KK-Ay mice
Compound Dose (mg/kg)
% Change
TG PG (V)LDL-C HDL-C
O
4b
14
1
3
3
ꢀ46 ꢀ12 ꢀ29
ꢀ56 ꢀ13 ꢀ49
ꢀ52 ꢀ18 ꢀ26
36
55
5
N
O
2
CO2H
O
R
Fenofibrate 300
ꢀ40 ꢀ25 ꢀ31
17
Compounds were orally administered to male KK-Ay mice (aged 10
weeks, n = 5) once a day for 4 days at the indicated dose.
Each value represents the percentage change of the mean value from
the control group. TG, plasma triglyceride; PG, plasma glucose;
(V)LDL-C, very-low-density plus low-density lipoprotein cholesterol;
HDL-C, high-density lipoprotein cholesterol.
Compound
R
Human PPAR subtype
(EC50, lM)a,b
a
c
d
4b
13
14
15
Me
H
0.3
100
(0%)
10
(0%)c
(3%)
(8%)
0.03
0.03
Et
i-Bu
(19%)
(6%)
10
a,b,c See corresponding footnotes to Table 1.
Compound 4b in particular, at a dose of 1 mg/kg, low-
ered TG to 46% of the levels of vehicle-treated animals,
although it was less potent in vitro than 14. The reason
for the apparent discrepancy between the in vitro and
in vivo results for these compounds is unclear, but it
may be due, for example, to differences in distribution
or metabolism between the two compounds. By con-
trast, fenofibrate 1a showed similar effects only at a
dosage as high as 300 mg/kg. These results provide
clear evidence that potent PPARa agonists can have
high efficacy in the treatment of hyperlipidemia and
hyperglycemia.
for mouse PPARa over mouse PPARc and d to justify
in vivo evaluation of their potential benefit (Table 4).
Each compound was orally administered to 10-week-
old male KK-Ay mice once a day for 4 days at 1–3
mg/kg. Both compounds markedly lowered plasma tri-
glyceride levels (TG) and very-low-density plus low-den-
sity lipoprotein cholesterol levels ((V)LDL-C)), while
they raised HDL cholesterol levels (HDL-C). These
compounds also decreased plasma glucose (PG) levels.

2132
T. Aoki et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2128–2132
Table 4. In vitro activity of 4b and 14 at mouse PPARa, b and d
subtypes
8. Xu, Y.; Mayhugh, D.; Saeed, A.; Wang, X.; Thompson,
R. C.; Dominianni, S. J.; Kauffman, R. F.; Singh, J.; Bean,
J. S.; Bensch, W. R.; Barr, R. J.; Osborne, J.; Montrose-
Rafizadeh, C.; Zink, R. W.; Yumibe, N. P.; Huang, N.;
Luffer-Atlas, D.; Rungta, D.; Maise, D. E.; Mantlo, N. B.
J. Med. Chem. 2003, 46, 5121.
9. Fujita, T.; Sugiyama, Y.; Taketomi, S.; Sohda, T.;
Kawamatsu, Y.; Suzuoki, Z. Diabetes 1983, 32, 804.
10. Sofia, M. J.; Chakravarty, P. K.; Katzenellenbogen, J. A.
J. Org. Chem. 1983, 48, 3318.
Compound
Mouse PPAR subtype (EC50, lM)a,b
a
c
d
4b
14
1
0.1
(21%)c
10
(6%)
(7%)
a,b,cSee corresponding footnotes to Table 1.
11. Godfrey, A. G.; Brooks, D. A.; Hay, L. A.; Peters, M.;
McCarthy, J. R.; Mitchell, D. J. Org. Chem. 2003, 68,
2623.
12. Moriya, T.; Seki, M.; Takabe, S.; Matsumoto, K.;
Takashima, K.; Mori, T.; Odawara, A.; Takeyama, S. J.
Med. Chem. 1988, 31, 1997.
In conclusion, we have synthesized a series of novel 1,3-
dioxane-2-carboxylic acid derivatives and evaluated
their PPAR subtype agonist activities. Structure-activity
relationship studies on the linker and the dioxane moiety
of this series revealed that the biological activity and
selectivity were sensitive to the length of the linker and
the geometrical configuration of the 1,3-dioxane ring.
This investigation led to the identification of several
compounds as highly potent and selective human
PPARa agonists, and the representative compound 4b
showed excellent hypolipidemic activity in diabetic
KK-Ay mice. Our results suggest that the development
of potent and selective PPARa agonists will be useful
in the treatment of hyperlipidemia and metabolic disor-
ders in type 2 diabetes.
13. Harabe, T.; Matsumoto, T.; Shioiri, T. Tetrahedron Lett.
2007, 48, 1443.
14. The configuration of the 1,3-dioxane ring moiety of 4b was
determined to be cis by 1H NMR analysis of the
corresponding alcohol 22, which has protons that allow
such a determination. The alcohol derivative 22 was
synthesized by LiAlH4 reduction of purified oxazole
methyl ester 10b free of trans-isomer. The a-protons at
the 4- and 6-positions were clearly distinguished from the
b-protons by the NOE correlation observed between the 5-
a-and 4-a(6-a)-protons. The strong NOE from the 4,6-b-
protons assigned above to the 2-hydroxymethyl protons
shows conclusively that the configuration is cis. The strong
NOE in 12b between the 4,6-b- and the 2-methyl protons
Acknowledgments
1
indicates a trans-configuration. H NMR of 22 (pyridine-
d5, 300 MHz) d1.25–1.30 (4H, m), 1.70 (2H, m), 1.71 (3H, s),
1.75 (1H, m, H-5a), 2.23 (3H, s), 2.48 (2H, t, J = 7.5 Hz),
3.72 (2H, dd, J = 11.7, 7.5 Hz, H-4,6b), 3.98 (2H, dd,
J = 11.7, 4.5 Hz, H-4,6a), 4.07 (2H, d, J = 4.2 Hz, 2b-
CH2OH), 7.21–7.58 (3H, m), 8.16–8.20 (2H, m). 1H NMR
of 12b (tetrahydrofuran-d8, 300 MHz) d 1.10–1.31 (2H, m),
1.36 (3H, s, 2b-CH3), 1.38–1.46 (1H, m, H-5a), 1.65–1.77
(4H, m), 2.31 (3H, s), 2.49 (2H, t, J = 7.5 Hz), 3.67 (2H, dd,
J = 11.7, 1.8 Hz, H-4,6a), 4.00 (2H, dd, J = 11.7, 1.5 Hz, H-
4,6b), 7.35–7.43 (3H, m), 7.93–7.96 (2H, m).
We thank Mr. Shoichi Chokai and Mr. Shinichi Tada
for practical guidance, and Dr. Akira Matsuura, Mr.
Masayuki Hattori and Dr. Gerald E. Smyth for helpful
suggestions during the preparation of the manuscript.
References and notes
1. (a) Kersten, S.; Desvergne, B.; Wahli, W. Nature 2000,
405, 421; (b) Berger, J.; Moller, D. E. Annu. Rev. Med.
2002, 53, 409.
2. (a) Braissant, O.; Foufelle, F.; Scotto, C.; Dauca, M.;
Wahli, W. Endocrinology 1996, 137, 354; (b) Auboeuf, D.;
Rieusset, J.; Fajas, L.; Vallier, P.; Frering, V.; Riou, J. P.;
Staels, B.; Auwerx, J.; Laville, M.; Vidal, H. Diabetes
1997, 46, 1319; (c) Staels, B.; Auwerx, J. Curr. Pharm. Des.
1997, 3, 1.
H
H
OH
Hβ
Hβ
CH3
2
Hβ
Hβ
6
6
2
OH
CH3
Hα
O
Hα
O
O
O
4
4
5
5
O
Hα
Hα
H
12b
H
22
3. Willson, T. M.; Brown, P. J.; Sternback, D. D.; Henke, B.
R. J. Med. Chem. 2000, 43, 527.
15. Shiotani, S.; Okada, H.; Nakamata, K.; Yamamoto, T.;
Sekino, F. Heterocycles 1996, 43, 1031.
16. Musso, D. L.; Cochran, F. R.; Kelley, J. L.; McLean, E.
W.; Selph, J. L.; Rigdon, G. C.; Orr, G. F.; Davis, R. G.;
Cooper, B. R.; Styles, V. L.; Thompson, J. B.; Hall, W. R.
J. Med. Chem. 2003, 46, 399.
17. Goto, Y.; Yamazaki, M.; Hamana, M. Chem. Pharm.
Bull. 1971, 19, 2050.
18. Glinski, M. B.; Durst, T. Can. J. Chem. 1983, 61, 573.
19. Charlton, J. L.; Alauddin, M. M. J. Org. Chem. 1986, 51,
3490.
4. (a) Torra, I. P.; Gervois, P.; Staels, B. Curr. Opin. Lipidol.
1999, 10, 151; (b) Seedorf, U.; Assmann, G. Nutr. Metab.
Cardiovasc. Dis. 2001, 11, 189.
5. Brown, P. J.; Winegar, D. A.; Plunket, K. D.; Moore, L.
B.; Lewis, M. C.; Wilson, J. G.; Sundseth, S. S.; Koble, C.
S.; Wu, Z.; Chapman, J. M.; Lehmann, J. M.; Kliewer, S.
A.; Willson, T. M. J. Med. Chem. 1999, 42, 3785.
6. Brown, P. J.; Stuart, L. W.; Hurley, K. P.; Lewis, M. C.;
Winegar, D. A.; Wilson, J. G.; Wilkison, W. O.; Ittoop, O.
R.; Willson, T. M. Bioorg. Med. Chem. Lett. 2001, 11,
1225.
20. Kuwabara, K.; Murakami, K.; Todo, M.; Aoki, T.; Asaki,
T.; Murai, M.; Yano, J. J. Pharmacol. Exp. Ther. 2004,
309, 970.
7. Miyachi, H.; Nomura, M.; Tanase, T.; Takahashi, Y.; Ide,
T.; Tsunoda, M.; Murakami, K.; Awano, K. Bioorg. Med.
Chem. Lett. 2002, 12, 77.