Efficient asymmetric synthesis of (S)-2-ethylphenylpropanoic acid derivative, a selective agonist for human peroxisome proliferator-activated receptor alpha.

Article abstract of DOI:10.1016/S0960-894X(02)00375-X

An optically active phenylpropanoic acid derivative, a selective agonist for human peroxisome proliferator-activated receptor alpha, was efficiently prepared in high optical purity by using Evans chiral oxazolidinone technique as a key step.

Full text of DOI:10.1016/S0960-894X(02)00375-X

Bioorganic & Medicinal Chemistry Letters 12 (2002) 2101–2104
Efficient Asymmetric Synthesis of (S)-2-Ethylphenylpropanoic
Acid Derivative, a Selective Agonist for Human Peroxisome
Proliferator-Activated Receptor Alpha
Masahiro Nomura, Takahiro Tanase and Hiroyuki Miyachi*
Discovery Research Laboratories, Kyorin Pharmaceutical Co., Ltd., 2399-1 Mitarai, Nogi-machi, Shimotsuga-gun,
Tochigi 329-0114, Japan
Received 15 April 2002; accepted 15 May 2002
Abstract—An optically active phenylpropanoic acid derivative, a selective agonist for human peroxisome proliferator-activated
receptor alpha, was efficiently prepared in high optical purity by using Evans chiral oxazolidinone technique as a key step. # 2002
Elsevier Science Ltd. All rights reserved.
Introduction
triglyceridemia for more than 20 years,8 and recently
fenofibrate (Chart 1) was launched in Japan. Although
the molecular target of these drugs remains to be defin-
itively established, recent molecular pharmacological
studies have demonstrated that fibrates activate PPARs
at high micromolar concentrations, with poor subtype
selectivity.9 We considered that more potent and selec-
tive activators of PPARa, especially human PPARa,
might have therapeutic utility for the treatment of
altered lipid homeostasis in the target organs, especially
in the liver. Recently, we have reported the design and
the synthesis of some novel phenylpropanoic acid deri-
vatives as subtype-selective PPAR agonists,10 and we
selected the 2-ethylphenylpropanoic acid derivative (4;
Chart 1) for further pharmacological study. We also
examined the enantio-dependency of 4 in peroxisome
proliferator-activated receptor a transactivation and
binding, and showed that these activities exclusively
reside on the (S)-isomer.11 In order to establish in detail
the in vivo pharmacological profile of (S)-(+)-4, and to
evaluate the compound as a candidate drugfor the
treatment of metabolic disorders, such as obesity, dia-
betes and others, a versatile asymmetric synthetic route
suitable to prepare large amounts was needed. In this
paper, we report an efficient asymmetric synthetic route
to (S)-(+)-4 in excellent enantiomeric excess.
Peroxisome proliferator-activated receptors (PPARs)
are members of the nuclear receptor family which
includes steroid receptor, thyroid receptor, retinoid
receptor and others.1 These receptors are ligand-depen-
dent transcription factors. Upon ligand binding, the
activated PPARs heterodimerize with another nuclear
receptor, the retinoid X receptor (RXR),2 and modulate
the transcription of the target genes after binding to
specific peroxisome proliferator response elements
(PPREs), which are a direct repeat of the hexameric
AGGTCA recognition motif separated by one nucleo-
tide (DR1).3 Three subtypes of PPARs, termed PPARa,
PPARd (also known as PPARb NUCI, FAAR) and
PPARg have been identified so far in various species,
4
includinghumans.
PPARa, the first isoform to be
cloned, in 1990,5 is present at high density in the liver
and regulates the expression of genes encoding proteins
involved in lipid and lipoprotein metabolism, such as
acyl-CoA oxidase, bifunctional enzyme, liver fatty acid
6
bindingprotein, apo A, apo C, and so on. In addition
to the above in vitro results, PPARa-deficient transgenic
mouse (PPARaÀ/À) exhibited massive hepatic and car-
diac lipid accumulation owingto inhibition of the cel-
lular fatty acid flux.7 These results clearly indicate a
pivotal role for PPARa in lipid homeostasis in vivo.
Fibrate compounds, such as clofibrate and bezafibrate
(Chart 1), have been used for the treatment of hyper-
Chemistry
As already reported, we have prepared (S)-(+)-4 by
optical resolution of the correspondingimide derivative
of 4, and subsequent removal of the chiral auxiliary.11
*Correspondingauthor. Tel.: +81-280-56-2201x286; fax: +81-280-
57-1293; e-mail: hiroyuki.miyachi@mb2.kyorin-pharm.co.jp
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00375-X

2102
M. Nomura et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2101–2104
However, the yield of (S)-(+)-4 was not high, and the
antipodal (R)-(À)-4 could not be racemized easily, so
this optical resolution method is not suitable from a
practical point of view.
(Chart 4).11 As already reported, (R)-4 that was pre-
pared from known (R)-(À)-BnBA exhibited levo rota-
tion. On the contrary, 4 prepared from the alkylated
product 6 exhibited antipodal dextro rotation, so the
absolute configuration of the C-2 ethyl group of 6 was
deduced to be (S).11
It was anticipated that the chiral a-ethylphenylpropa-
noic acid derivative 4 could be prepared by usingEvans
asymmetric alkylation methodology,12 with subsequent
derivatization (Chart 2). Therefore, to test this idea, we
first examined the reaction between the acyl oxazolidi-
We next examined the effects of the organic base and
the reaction temperature, since a very low temperature
reaction condition (such as À78 ꢀC) is not convenient
from a practical point of view. At a higher temperature
of about À30 ꢀC, no reaction took place and the starting
materials were recovered. The reason for this is prob-
ably the instability of the sodium enolate of the oxazo-
lidinone formed in the medium.12 Then we tried the
reaction at À30 ꢀC usinglithium hexamethyldisilazide
(LiHMDS) instead of NaHMDS, because the corre-
spondinglithium enolate derivative of the oxazolidi-
none was expected to be more stable at higher
temperature than the sodium enolate derivative.12 As
expected, the reaction proceeded smoothly and gave 6 in
none derivative
5
and benzyl 5-bromomethyl-2-
methoxybenzoate (which was prepared from methyl 5-
formyl-2-methoxy benzoate in four steps) at À78 ꢀC,
usingan equimolar amount of sodium hexamethyldisil-
azide (NaHMDS) as a base.13 The reaction was com-
plete in about 10 h and gave the desired alkylated
product 6 in about 64% yield, with 97% diastereomeric
excess (see Chart 3).14 The absolute configuration of the
C-2 ethyl group was determined by comparison with the
final product 4, which was prepared both from 6 and
from (R)-(À)-2-benzylbutanoic acid ((R)-(À)-BnBA)
Chart 1. Chemical structures of the known fibrate drugs and 4.
Chart 2. Retro synthetic scheme for (S)-(+)-4.
Chart 3. Synthetic route to (S)-(+)-4. Reagents and conditions: (a) LiHMDS, benzyl 5-bromomethyl-2-methoxybenzoate, THF, À20 ꢀC, 5 h, 80%;
.
(b) H2, 10% Pd/C, EtOH, quant; (c) ClCO2Et, TEA, 4-(trifluoromethyl)benzylamine, CH2Cl2, 90%; (d) LiOH H2O, 30% H2O2, 87%.

M. Nomura et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2101–2104
2103
Chart 4. Synthetic route to (R)-(À)-4 startingfrom known ( R)-(À)-BnBA.
about 80% yield, with 98% enantiomeric excess. When
the reaction was performed at 0 ꢀC, the reaction was
incomplete and only a 20% yield of 6 was obtained.
Benzyl 5-bromomethyl- 2-methoxybenzoate was impor-
tant, because no alkylated product was obtained when
the correspondingchloromethyl derivative was used as
the electrophile, probably due to insufficient reactivity
of the electrophile.
discussions. Thanks are also due to H. Saito, H. Furuta,
S. Isogai, and H. Kobayashi of Kyorin Pharmaceutical
Co., Ltd., for their technical work.
References and Notes
1. Porte, D., Jr.; Schwartz, M. W. Science 1996, 272, 699.
2. Keller, H.; Dreyer, C.; Medin, J.; Mahoudi, A.; Ozato, K.;
Wahli, W. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 2160.
3. Kliewer, S. A.; Umesono, K.; Noonan, D. J.; Heyman,
R. A.; Evans, R. M. Nature (Lond.) 1992, 358, 771.
4. Staels, B.; Auwerx, J. Curr. Pharmaceut. Des 1997, 3, 1.
5. Isseman, I.; Green, S. Nature (Lond.) 1990, 347, 771.
6. Staels, B.; Dallongeville, J.; Auwerx, J.; Schoonjans, E.;
Leitersdorf, E.; Fruchart, J.-C. Circulation 1998, 98, 2088.
7. Lee, S. S.; Pineau, T.; Drago, J.; Lee, E. L.; Owens, J. W.;
Kroetz, D. L.; Fernandez-Salguero, P. M.; Westphal, H.;
Gonzalez, F. J. Mol. Cell. Biol. 1995, 15, 3012.
8. Fruchart, J.-C.; Duriez, P.; Staels, B. Curr. Opin. Lipidol.
1999, 10, 245.
9. Forman, B. M.; Chen, J.; Evans, R. M. Proc. Natl. Acad.
Sci. U.S.A. 1997, 94, 4312.
10. (a) Nomura, M.; Takahashi, Y.; Tanase, T.; Miyachi, H.;
Ide, T.; Tsunoda, M.; Murakami, K.; PCT Int. Appl. WO 00/
75103. (b) Miyachi, H.; Nomura, M.; Tanase, T.; Takahashi,
Y.; Ide, T.; Tsunoda, M.; Murakami, K.; Awano, K. Bioorg.
Med. Chem. Lett. 2002, 12, 77.
11. Miyachi, H.; Nomura, M.; Tanase, T.; Takahashi, Y.; Ide,
T.; Tsunoda, M.; Murakami, K.; Awano, K. Bioorg. Med.
Chem. Lett. 2002, 12, 333.
Consideringthese preliminary results, the synthesis of
(S)-(+)-4 was performed as follows (Chart 3). (R)-N-
Butyryl-4-benzyloxazolidinone 5 was treated with ben-
zyl 5-bromomethyl- 2-methoxybenzoate at À20ꢀC for 5 h
in the presence of an equimolar amount of LiHMDS as
a base to afford 90% yield of the desired alkylated
product 6, with high enentiomeric excess (98% e.e.).
Subsequent hydrogenolysis afforded the important ver-
satile synthetic intermediate 7 almost quantitatively.
4-(Trifluoromethyl)benzylamine was condensed with 7
by the mixed anhydride method (90% yield), followed
by the removal of the chiral auxiliary of 8 by alkaline
hydrolysis to afford the desired (S)-(+)-4 with high
retention of enantiomeric excess (98% e.e.).15
In conclusion, we have developed an efficient and prac-
tical asymmetric synthetic route to an optically active 2-
ethylphenylpropanoic acid derivative (S)-(+)-4, using
Evans’s asymmetric alkylation methodology as a key
step. Since (S)-(+)-4 is a very potent and subtype-
selective human PPARa agonist,10,11 it not only repre-
sents a useful pharmacological tool to investigate the
physiology and pathophysiology of PPARa, but is also
a candidate drugfor the clinical treatment of metabolic
disorders, such as dyslipidemia, obesity, and diabetes
(Fig. 1).
12. Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem.
Soc. 1982, 104, 1737.
13. (R)-3-(1-Butyryl)-4-benzyloxazolidin-2-one (3.37 g, 13.6 mmol)
and 70 mL of dehydrated tetrahydrofuran were mixed under
an atmosphere of argon, and cooled to À78 ꢀC. Under stirring,
a
1 mol/L solution of NaHMDS in dehydrated tetra-
hydrofuran (15.0 mL, 15.0 mmol) was added dropwise. After
completion of the addition, the mixture was stirred for 1 h at
À78 ꢀC and then a solution of benzyl 5-bromomethyl-2-
methoxybenzoate (5.04 g, 15.0 mmol) in dehydrated tetra-
hydrofuran (20 mL) was added dropwise. After the addition,
the mixture was stirred for 6 h at À78 ꢀC. Saturated aqueous
ammonium chloride was added, and the whole was extracted
with ethyl acetate. The organic solution was washed with
water and brine, dried over anhydrous sodium sulfate, and
concentrated. The residue was purified by silica gel column chro-
matography (eluant; n-hexane/ethyl acetate=4:1v/v) to obtain
4.38 g(64%) of 6 as a colorless oil; 1H NMR (400 MHz, CDCl3) d
0.93 (3H, t, J=7.3 Hz), 1.51–1.63 (1H, m), 1.71–1.82 (1H, m),
2.43 (1H, dd, J=13.2, 9.8 Hz), 2.75 (1H, dd, J=13.7, 6.3 Hz),
2.99–3.08 (2H, m), 3.86 (3H, s), 4.03–4.15 (3H, m), 4.64 (1H, m),
5.27 (1H, d, J=12.2 Hz), 5.31 (1H, d, J=12.7Hz), 6.91 (1H, d,
J=8.8Hz), 7.03 (2H, dd, J=7.8, 2.0Hz), 7.20–7.42 (9H, m), 7.72
(1H, d, J=2.0 Hz); low-resolution MS (EI+) m/e 501(M+).
14. Analysis of the enantiomeric excess was performed by HPLC
with a CHIRAPAC OD column (0.0046Â0.25 m, flow rate
1.00 mL/min, UV 254 nm, n-hexane/i-PrOH: TFA=95:5:0.2 v/
v/v as the eluant).
Figure 1. A PPARa selective agonist (S)-(+)-4.
Acknowledgements
The authors wish to thank to Dr. K. Murakami, man-
ager of Kyorin Pharmaceutical Co., Ltd., for his helpful

2104
M. Nomura et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2101–2104
15. Physicochemical properties of (S)-(+)-4 were as follows;
mp 128–130 ꢀC; 1H NMR (400 MHz, CDCl3) d 0.95 (3H, t,
J=7.3 Hz), 1.54–1.69 (2H, m), 2.58–2.65 (1H, m), 2.77 (1H,
dd, J=13.7, 6.3 Hz), 2.96 (1H, dd, J=13.7, 8.3 Hz), 3.92 (3H,
s), 4.38 (1H, br s), 4.72 (2H, d, J=5.9 Hz), 6.90 (1H, d,
J=8.3 Hz), 7.29 (1H, dd, J=8.3, 2.4 Hz), 7.46 (2H, d,
J=7.8 Hz), 7.58 (2H, d, J=7.8 Hz), 8.07 (1H, d, J=2.4 Hz),
8.34 (1H, t, J=5.9 Hz); low-resolution MS (EI+) m/e 451 (M+);
[a]D +25ꢀ (c 0.8, MeOH). Anal. calcd for C21H22F3NO4: C,
61.61; H, 5.42; N, 3.42. Found: C, 61.41; H, 5.44; N, 3.41.