Concise and efficient asymmetric synthesis of (S)-2-ethylphenylpropanoic acid derivatives: Dual agonists for human peroxisome proliferator-activated receptor α and δ

Article abstract of DOI:10.1016/j.bmcl.2005.11.029

Optically active (S)-2-ethylphenylpropanoic acid derivatives, dual agonists for human peroxisome proliferator-activated receptor (PPAR) α and δ, were efficiently prepared by using Evan's chiral oxazolidinone technique and reductive amide N-alkylation as k

Full text of DOI:10.1016/j.bmcl.2005.11.029

Bioorganic & Medicinal Chemistry Letters 16 (2006) 771–774
Concise and efficient asymmetric synthesis of
(S)-2-ethylphenylpropanoic acid derivatives: Dual agonists
for human peroxisome proliferator-activated receptor a and d
Jun-ichi Kasuga, Yuichi Hashimoto and Hiroyuki Miyachi*
Institute of Molecular and Cellular Biosciences, University of Tokyo, Japan
Received 17 October 2005; revised 6 November 2005; accepted 8 November 2005
Available online 29 November 2005
Abstract—Optically active (S)-2-ethylphenylpropanoic acid derivatives, dual agonists for human peroxisome proliferator-activated
receptor (PPAR) a and d, were efficiently prepared by using EvansÕs chiral oxazolidinone technique and reductive amide N-alkyl-
ation as key steps.
Ó 2005 Elsevier Ltd. All rights reserved.
Peroxisome proliferator-activated receptors (PPARs)
are members of the nuclear receptor family, which
includes steroid, thyroid, retinoid, vitamin D, and other
receptors.1 To date, three subtypes of PPARs, termed
PPARa [officially NR1C1], PPARd (also known as
PPARb, NUCI, FAAR) [NR1C2], and PPARc
[NR1C3], have been identified in various species, includ-
ing humans.2 Each PPAR subtype appears to be differ-
entially expressed in a tissue-specific manner and to
play a pivotal role in lipid, lipoprotein, and glucose
homeostasis.3 PPARa is mostly expressed in the tissues
involved in lipid oxidation, such as liver, kidney, skeletal
muscle, cardiac muscle, and adrenal glands. PPARc is
expressed in adipose tissue, macrophages, and vascular
smooth muscles. In contrast to the specific distributions
of PPARa and PPARc, PPARd is ubiquitously
expressed in almost all mammalian tissues.
PPARa regulates the expression of genes encoding for
proteins involved in lipid and lipoprotein homeostasis.5
For example, it regulates genes involved in fatty acid up-
take (such as fatty acid binding protein, FABP), b-oxi-
dation
(acyl-CoA
oxidase),
and
x-oxidation
(cytochrome P450). It down-regulates apolipoprotein
C-III, a protein that inhibits triglyceride hydrolysis by
lipoprotein lipase,6 and it also regulates genes involved
in reverse cholesterol transport, such as apolipoprotein
A-I and apolipoprotein A-II.5
Based on the findings that antidiabetic thiazolidine-
2,4-diones (glitazones), and antidyslipidemic fibrates
are ligands of PPARc and PPARa, respectively, much
research has been focused on these metabolic nuclear
receptor subtypes as therapeutic targets for the treat-
ment of metabolic syndrome, and compounds such
as GW-9578 (4)6 and KCL (5)7 (Fig. 1) have been
developed.
PPARc was first identified as a master regulator of adi-
pocyte differentiation, but more recent molecular–bio-
logical studies have indicated that PPARc activation is
also linked to the expression of many important genes
that affect energy production, such as the TNF-a, leptin,
and adiponectin genes.4
Although PPARd is ubiquitously distributed in a wide
range of tissues and cells, research interest in PPARd
has been limited. However, the availability of selective
ligands, such as GW-501516 (7)8 and L-165041 (8)9
(Fig. 1), prompted us to examine the roles of PPARd
in fatty acid metabolism, reverse cholesterol transport,
and other disease states. Considering the positive contri-
butions of PPARa and PPARd to lipid, lipoprotein, and
cholesterol homeostasis, dual agonists of both PPARa
and PPARd might be candidates of choice for the treat-
ment of metabolic syndrome, by decreasing triglyceride,
Keywords: PPARa; PPARd; Metabolic syndrome; Reductive amide
alkylation.
*
Corresponding author. E-mail: miyachi@iam.u-tokyo.ac.jp
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.11.029

772
J. Kasuga et al. / Bioorg. Med. Chem. Lett. 16 (2006) 771–774
O
O
O
O
O
S
F
OH
O
N
NH
NH
Me
Me
Me
NH
S
H
N
N
S
N
H
N
F
F
S
O
MeO
N
O
O
F
O
O
F
1
2
3
4
O
O
Me
O
Me
O
F
Cl
O
O
O
OH
OH
N
OH
O
O
H
OH
Cl
F
F
F
F
H
N
OH
Me
S
MeO
S
Cl
N
Me Me
F
F
O
O
8
N
O
Me
7
5
6
Figure 1. Structures of representative PPARc (1 (pioglitazone), 2 (rosiglitazone)), dual PPARa/c (3 (KRP-297/MK767), PPARa (4 (GW-9578), 5
(KCL)), dual PPARa/d (6 (GW-2433)), and PPARd (7 (GW-501516), 8 (L-165041)) agonists.
free fatty acid, LDL-cholesterol, and increasing HDL
cholesterol. But, there have been few reports of such
agonists, except for GW-2433 (6) (Fig. 1), and its poten-
cy is not high.
optically active benzylamine derivative (11) (Scheme
1). Compound 11 was expected to be obtainable from
the corresponding benzoic acid derivative (12), used
for the synthesis of optically active human PPARa selec-
tive agonist KCL (5).7 However, several attempts failed
to afford the desired benzylamine derivative (11) in good
yield with high purity.
Recently, we reported the design and synthesis of novel
phenylpropanoic acid derivatives as dual PPARa/d ago-
nists,10 and we selected the 2-ethylphenylpropanoic acid
derivatives (9a,b; Fig. 2) for further pharmacological
study. We also discovered that the agonistic activity de-
pends on the stereochemistry of the a-substituted ethyl
group, and the (S)-form is more potent than the anti-
pode.10 In order to establish in detail the in vivo pharma-
cological profiles of 9a,b, and to evaluate the compounds
as candidate drugs for the treatment of metabolic syn-
drome, a versatile asymmetric synthetic route suitable
to prepare large amounts of 9a,b was needed. In this pa-
per, we report an efficient asymmetric synthetic route to
9a,b in excellent enantiomeric excess (Scheme 1).
We then planned the synthesis of 9a,b by the N-alkylation
of 2-fluoro-4-trifluoromethylbenzamide (or 3-fluoro-4-
trifluoromethylbenzamide) withthe bromomethyl deriva-
tive (19) (Scheme 3 step b). Compound 19 was prepared
by the reduction of 12 with BH3–tetrahydrofuran com-
plex and subsequent bromination with PPh3–CBr4 (yield
67%, two steps).12 3-Fluoro-4-trifluoromethylbenzamide
was treated with NaH (or LiHMDS, or t-BuOK) and then
treated with 19 to afford several compounds, including 20.
After silica gel column chromatography, 20 was isolated
in only 15% yield, and a considerable amount of the start-
ing 3-fluoro-4-trifluoromethylbenzamide was recovered.
Therefore, direct N-alkylation was also found to be
unsuitable in this case.
We previously prepared 9a and b by the method depict-
ed in Scheme 2.11 However, this route contains multiple
steps with chromatographic purification of intermedi-
ates, and a low total yield (seven steps, 3% overall yield),
and it is unsuitable for scale-up.
Recently, Dobe and Scholte reported an efficient reduc-
tive N-alkylation of amides using TFA/Et3SiH with
aldehyde.13 Since the aldehyde (21) was readily prepared
by the reduction of 12, followed by oxidation with acti-
vated MnO2 (90% yield, two steps),14 we examined the
application of this methodology to the synthesis of
9a,b (Scheme 3 step e).
In our initial design, we planned the synthesis of 9a,b by
the condensation of 2-fluoro-4-trifluoromethylbenzoic
acid (or 3-fluoro-4-trifluoromethylbenzoic acid) and an
O
O
When a mixture of 2-fluoro-4-trifluoromethylbenzamide
(1 equiv), the aldehyde (21) (3 equiv), triethylsilane
(3 equiv), and trifluoroacetic acid (3 equiv) in toluene
was heated to reflux for 24 h, the desired N-alkylation
product (20) was successfully obtained in good (66%)
yield after silica gel column chromatography.15 After
deprotection of the chiral oxazolidinone moiety with
the LiOH/30% H2O2 system,16 the desired 9a was
X
OH
N
H
F
F
MeO
F
9
a; X = 2-F
b; X = 3-F
Figure 2. Structures of dual PPARa/d agonists 9a and 9b.
F
O
O
O
O
O
O
O
O
F
O
O
N
N
H2N
MeO
N
HO
MeO
N
O
O
O
H
N
OH
F
F
MeO
H
F
F
MeO
F
F
12
9a
10
11
Scheme 1. Retrosynthetic analysis of the preparation of 9a.

J. Kasuga et al. / Bioorg. Med. Chem. Lett. 16 (2006) 771–774
773
O
O
O
O
O
O
O
HO
N
HO
N
TBDMSO
MeO
N
O
O
O
O
d
MeO
MeO
a
b
c
TBDMSO
MeO
O
F
15
12
13
14
F
O
O
O
O
O
O
O
OH
N
N
g
e
f
HO
O
H2N
O
H
H
F
F
F
F
MeO
MeO
MeO
MeO
F
F
16
17
18
9a
Scheme 2. Initial synthetic route to 9a. Reagents and conditions: (a) BH3–tetrahydrofuran, THF, 0 °C, overnight, 90%; (b) TBDMSCl, imidazole,
THF, rt, overnight, 65%; (c) benzyl alcohol, n-BuLi in tetrahydrofuran, THF, À40 °C, 4 h, quant.; (d) TBAF, THF, rt, overnight, 87%; (e) (1) P(Ph)3-
polymer, CBr4, rt, overnight, 86%, (2) potassium phthalimide, DMF, 100 °C, 2 h, 80%, (3) NaBH4, iPrOH–H2O, rt, overnight then AcOH, 80 °C,
3 h, 37%; (f) ethyl chloroformate, triethylamine, 0 °C, 20 min, then 2-fluoro-4-trifluoromethylbenzoic acid, THF, rt, overnight, 26%; (g) H2, 10% Pd-
C, AcOEt, rt, overnight, 60%.
O
O
O
F
O
O
O
O
O
F
O
N
Br
N
N
N
HO
O
O
O
H
F
F
OH
N
MeO
MeO
a
b
c
MeO
H
F
F
F
MeO
F
13
19
20
9a
O
O
O
N
H
O
d
e
MeO
21
Scheme 3. Preparation of 9a by amide N-alkylation or reductive amide N-alkylation. Reagents and conditions: (a) P(Ph)3-polymer, CBr4, rt
overnight, 70%; (b) (1) 60% NaH, THF, 0 °C, then 2-fluoro-4-trifluoromethylbenzamide, THF, rt, overnight, 15%; (c) LiOH H2O, 30% H2O2, THF/
H2O = 4:1 (v/v), 85%; (d) activated MnO2, rt, quant.; (e) 2-fluoro-4-trifluoromethylbenzamide, triethylsilane, trifluoroacetic acid, toluene, reflux,
66%.
PPARd, but also candidate drugs for the clinical treat-
ment of metabolic syndrome.
Acknowledgments
The work described in this paper was partially support-
ed by Grants-in-Aid for Scientific Research from The
Ministry of Education, Culture, Sports, Science and
Technology, Japan, and the Japan Society for the
Promotion of Science.
Figure 3. Dose–response of the transactivation of PPARa/d agonists
9a.
References and notes
1. Porte, D., Jr.; Schwartz, M. W. Science 1996, 272, 699.
2. Staels, B.; Auwerx, J. Curr. Pharm. Des. 1997, 3, 1.
3. Willson, T. M.; Brown, P. J.; Sternbach, D. D.; Henke, B.
R. J. Med. Chem. 2000, 43, 527.
obtained in 85% yield. Namely, 9a was obtained from 12
in a total yield of about 50% in only three steps. Similar-
ly, 9b was obtained from 12 in total yield of about 50%.
4. Okuno, A.; Tamemoto, H.; Tobe, K.; Ueki, K.; Mori, Y.;
Iwamoto, K.; Umesono, K.; Akanuma, Y.; Ujiwara, T.;
Horikoshi, H.; Yazaki, Y.; Kadowaki, T. J. Clin. Invest.
1998, 101, 1354.
5. Mukherjee, R.; Jow, L.; Noonan, D.; McDonnell, D. P. J.
Steroid Biochem. Mol. Biol. 1994, 51, 157.
6. 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.
In conclusion, we have developed an efficient and prac-
tical synthetic route to the optically active 2-ethyl-
phenylpropanoic acid derivatives 9a and b, using
EvansÕs asymmetric alkylation and reductive N-alkyl-
ation as key steps. Since 9a and b are very potent and
human PPARa/d dual agonists (Fig. 3),17 they should
not only be useful pharmacological tools to investigate
the physiology and pathophysiology of PPARa and

774
J. Kasuga et al. / Bioorg. Med. Chem. Lett. 16 (2006) 771–774
7. (a) Nomura, M.; Takahashi, Y.; Tanase, T.; Miyachi, H.;
13. Dube, D.; Scholte, A. A. Tetrahedron Lett. 1999, 40, 2295.
14. A mixture of 13 (1.50 g, 3.89 mmol), activated MnO2
(3.00 g, 42.3 mmol), and 40 mL of dehydrated dichloro-
methane was stirred overnight at rt. The mixture was
filtered through Celite and the filtrate was evaporated. The
residue was purified by silica gel column chromatography
(eluant; n-hexane/ethyl acetate 3:2 v/v) to obtain 1.50 g
(quant.) of 21 as a yellow oil; 1H NMR (500 MHz, CDCl3)
d 0.91 (3H, t, J = 7.3 Hz), 1.51–1.55 (1H, m), 1.72–1.78
(1H, m), 2.51 (1H, dd, J = 13.3, 6.8 Hz), 2.74 (1H, dd,
J = 6.8, 6.8 Hz), 3.03 (1H, dd, J = 13.3, 7.3 Hz), 3.09 (1H,
dd, J = 13.3, 3.0 Hz), 3.87 (3H, s), 4.01–4.15 (3H, m), 4.64
(1H, m), 6.90 (1H, d, J = 8.6 Hz), 7.05 (2H, d, J = 6.4 Hz),
7.22–7.27 (2H, m), 7.50 (1H, dd, J = 8.6, 2.6 Hz), 7.67
(1H, d, J = 2.6 Hz), 10.4 (1H, s); [a]D À68.1 (c = 0.356,
MeCN).
Ide, T.; Tsunoda, M.; Murakami, K.; PCT Int. Appl. WO
00/75103; (b) Nomura, M.; Tanase, T.; Ide, T.; Tsunoda,
M.; Suzuki, M.; Uchiki, H.; Murakami, K.; Miyachi, H. J.
Med. Chem. 2003, 46, 3581.
8. Sznaidman, M. L.; Haffner, C. D.; Maloney, P. R.;
Fivush, A.; Chao, E.; Goreham, D.; Sierra, M. L.;
LeGrumelec, C.; Xu, H. E.; Montana, V. G.; Lambert,
M. H.; Willson, T. M.; Oliver, W. R., Jr.; Sternbach, D. D.
Bioorg. Med. Chem. Lett. 2003, 13, 1517.
9. Lim, H.; Gupta, R. A.; Ma, W. G.; Paria, B. C.; Moller,
D. E.; Morrow, J. D.; DuBois, R. N.; Trzaskos, J. M.;
Dey, S. K. Genes Dev. 1999, 13, 1561.
10. Kasuga, J.; Makishima, M.; Hashimoto, Y.; Miyachi, H.
Bioorg. Med. Chem. Lett. 2005, accepted for publication.;
11. Nomura, M.; Tanase, T.; Miyachi, H. Bioorg. Med. Chem.
Lett. 2002, 12, 2101.
15. A mixture of 21 (750 mg, 1.95 mmol), 2-fluoro-4-trifluo-
romethylbenzamide (1.21 g, 5.85 mmol), triethylsilane
(0.94 mL, 5.85 mmol), trifluoroacetic acid (0.45 mL,
5.85 mmol), and 30 mL of dehydrated toluene was
refluxed for 24 h. The mixture was evaporated, and the
residue was purified by silica gel column chromatography
(eluant; n-hexane/ethyl acetate 2:1 v/v) to obtain 750 mg
(66%) of 20 as a colorless oil; 1H NMR (500 MHz, CDCl3)
d 0.93 (3H, t, J = 7.6 Hz), 1.56—1.60 (1H, m), 1.73—1.79
(1H, m), 2.44 (1H, dd, J = 13.7, 6.4 Hz), 2.76 (1H, dd,
J = 13.7, 6.4 Hz), 2.94–3.02 (2H, m), 3.86 (3H, s), 4.03
(1H, dd, J = 9.0, 2.6 Hz), 4.09–4.12 (2H, m), 4.57–4.69
(3H, m), 6.84 (1H, d, J = 8.1 Hz), 6.94 (2H, s), 7.18–7.31
(6H, m), 7.46 (1H, d, J = 8.1 Hz), 8.15 (1H, t, J = 8.1 Hz);
[a]D À58.3 (c = 0.114, MeCN).
12. A mixture of 12 (1.20 g, 2.90 mmol) was dissolved in
30 mL of dehydrated tetrahydrofuran and cooled to 0 °C.
3.50 mL of 1 mol/L borane–tetrahydrofuran complex was
added dropwise and stirred overnight at rt. The reaction
mixture was poured into satd NH4Cl solution and
extracted with ethyl acetate. The extract was dried,
filtered, and evaporated. The residue was purified by silica
gel column chromatography (eluant; n-hexane/ethyl
acetate 1:1 v/v) to obtain 1.10 g (95%) of 13 as a colorless
oil; 1H NMR (500 MHz, CDCl3) d 0.99 (t, J = 7.3 Hz,
3 H), 1.65 (m, 2H), 1.82 (dd, J = 13.3, 9.4 Hz, 1H), 2.33
(m, 1H), 2.56 (dd, J = 13.3, 6.4 Hz, 1H), 2.80 (dd, J = 13.3,
6.9 Hz, 1H), 3.07 (dd, J = 13.3, 6.9 Hz, 1H), 3.14 (dd,
J = 13.3, 6.9 Hz, 1H), 3.89 (s, 3H), 4.13–4.22 (m, 3H), 4.71
(d, J = 6.0 Hz, 2H), 6.86 (d, J = 8.5 Hz, 1H), 7.12 (m, 2H),
7.24–7.34 (m, 5H); MS (FAB+) 397 (M); [a]D À36.3
(c = 0.52, MeCN). A mixture of 13 (958 mg, 2.41 mmol),
P(Ph)3 (840 mg, 3.20 mmol), CBr4 (1.06 g, 3.20 mmol),
and 25 mL CH2Cl2 was stirred for overnight at rt. The
reaction solvent was evaporated and the residue was
purified by silica gel column chromatography (eluant;
n-hexane/ethyl acetate 2:1 v/v) to obtain 720 mg (70%) of
19 as a colorless oil; 1H NMR (500 MHz, CDCl3) d 0.94 (t,
J = 7.3 Hz, 3H), 1.57 (m, 1H), 1.77 (m, 1H), 2.41 (dd,
J = 13.3, 9.4 Hz, 1H), 2.75 (dd, J = 13.6, 6.4 Hz, 1H), 2.97
(dd, J = 13.6, 6.4 Hz, 1H), 3.03 (m, 1H), 3.84 (s, 3H), 4.05
(m, 1H), 4.12 (m, 2H), 4.52 (s, 2H), 4.65 (m, 1H), 6.78 (d,
J = 8.5 Hz, 1H), 7.05 (d, J = 8.5 Hz, 2H), 7.20 (m, 5H);
HRMS (EI+) calcd for C23H27BO479Br (M + H) 460.1132,
found 460.1161. This compound was used without further
purification.
16. Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem.
Soc. 1982, 104, 1737.
17. Human embryonic kidney HEK293 cells were cultured in
DMEM containing 5% fetal bovine serum and antibiotic–
antimycotic at 37 °C in a humidified atmosphere of 5%
CO2 in air. Transfections were performed by the calcium
phosphate coprecipitation. Eight hours after transfection,
ligands were added. Cells were harvested approximately
16–20 h after the treatment, and luciferase and b-galacto-
sidase activities were assayed using a luminometer and a
microplate reader. DNA cotransfection experiments
included 50 ng reporter plasmid, 20 ng pCMX-b-galacto-
sidase, 15 ng of each PPAR receptor, and pGEM carrier
DNA for a total of 150 ng DNA per well in a 96-well
plate. Luciferase data were normalized to an internal
b-galactosidase control and represent means ( SD) of
triplicate assays.