A refined method for determining the absolute configuration of the 3-hydroxy-3-methylglutaryl group

Article abstract of DOI:10.1016/j.tetasy.2007.05.013

A refined method for the determination of the absolute configuration of the 3-hydroxy-3-methylglutaryl group in natural products was developed. An interchange between the order of reduction with lithium borohydride and amidation with a chiral amine descri

Full text of DOI:10.1016/j.tetasy.2007.05.013

Tetrahedron: Asymmetry 18 (2007) 1183–1186
A refined method for determining the absolute configuration
of the 3-hydroxy-3-methylglutaryl group
Yasunao Hattori,a,b Ko-hei Horikawa,c Hidefumi Makabe,d Nobuhiro Hirai,e
Mitsuru Hirotac and Tsunashi Kamoc,*
aInterdisciplinary Graduate School of Science and Technology, Shinshu University, 8304, Minami-minowa,
Kami-ina, Nagano 399-4598, Japan
bSatellite Venture Business Laboratory, Shinshu University, 3-15-1, Tokida, Ueda, Nagano 386-8567, Japan
cDepartment of Bioscience and Biotechnology, Faculty of Agriculture, Shinshu University, 8304 Minami-minowa,
Kami-ina, Nagano 399-4598, Japan
dSciences of Functional Foods, Graduate School of Agriculture, Shinshu University, 8304 Minami-minowa,
Kami-ina, Nagano 399-4598, Japan
eInternational Innovation Center, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan
Received 28 March 2007; accepted 9 May 2007
Abstract—A refined method for the determination of the absolute configuration of the 3-hydroxy-3-methylglutaryl group in natural
products was developed. An interchange between the order of reduction with lithium borohydride and amidation with a chiral amine
described in the previous procedure resulted in a high yield of the desired derivative. The refined method is applicable to various natural
products with the 3-hydroxy-3-methylglutaryl group, even if their available amounts are 0.1 mg. The absolute configuration of the group
in the lanostane-type triterpene acid, which had been isolated in a small amount, was established as S.
ꢀ 2007 Elsevier Ltd. All rights reserved.
26
CO2H
1. Introduction
HO
The 3-hydroxy-3-methylglutaryl (HMG) group is occasion-
ally found in natural products, such as terpenoids,1–16
flavonoids,17,18 and phenylpropanoids.19–24 We have previ-
ously isolated a known lanostane-type triterpene acid with
HMG group 1 and a new one 2 from the fruiting bodies of
Piptoporus betulinus and found their anti-inflammatory
activities (Fig. 1).1
OH
O
HO2C
O
3'
5'
R
1: R=CH3
2: R=CH2OH
Figure 1. Lanostane-type triterpene acids from Piptoporus betulinus.
Naturally occurring HMG esters, which are formed via the
acylation of the hydroxy group with (S)-HMG-CoA, pos-
sess an (S)-configuration at the C-3 stereogenic carbon.25
The absolute configuration of the HMG groups in natural
products has been confirmed by amidation with L-alanine
methyl ester6 or reduction with LiBH4 to mevalonolac-
tone.2–5 We used the LiBH4 reduction to establish the con-
figuration of 1 as (S).1 This reagent is more suitable for the
selective reduction of the ester when compared to other
hydride donors such as LiAlH4 and NaBH4: the former
reduces not only esters but also carboxylic acids, while
the latter requires high temperature to reduce esters.
Although the LiBH4 reduction is a proper method for the
chemical determination of the absolute configuration of
the HMG group, the reduction and following acid-
catalyzed cyclization gave mevalonolactone in low yields
(10–30%), requiring large amounts of natural products
(11–300 mg, 11–250 lmol).1–5 We were unable to determine
*
Corresponding author. Tel.: +81 265 77 1602; e-mail: kamo274@
shinshu-u.ac.jp
0957-4166/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.05.013

1184
Y. Hattori et al. / Tetrahedron: Asymmetry 18 (2007) 1183–1186
the absolute configuration of 2 due to the small amount
available (4.3 mg).1 Thus, the absolute configuration of
the HMG groups in natural products has been sometimes
left unknown.12–21
step without purification after amidation, indicating that
the method was applicable to a wide variety of natural
products with the HMG group. Although alcohol 4 needed
to be purified before acetylation in the case of a small-
scaled derivatization, (3R)-4 and (3S)-4 were eluted at the
same retention time on an ODS column, which was con-
firmed by purification of the diastereomers derived from
commercial racemic mevalonolactone (data not shown).
In total, amide 5 was prepared from 1 through three steps
in a 35% yield. Thus, 0.1 mg (0.16 lmol) of 1 gave enough
sample to be analyzed by HPLC for determining the abso-
lute configuration of the HMG group (Fig. 2). Compared
to the determination process previously reported, the scale
down of the starting ester amount to smaller than 1% was
achieved by the present scheme. For HPLC analysis, ca.
10 lg of amide 5 including co-chromatography is required.
Although the required amount of the starting material
depends on the yield of the LiBH4 reduction, the minimal
amount would be ca. 0.1 mg in many cases.
Herein we report a refined method for the determination of
the absolute configuration of the HMG group, which is
applicable to a small amount of natural products. By using
this method, we have determined the absolute configura-
tion of the HMG group of 2.
2. Results and discussion
We interchanged the order of reduction and amidation
described in the previous method (Scheme 1). As a model
compound, we used 1, the absolute configuration of which
has already been established as (S). Amidation of 1 (23 mg)
with (S)-(ꢀ)-phenylethylamine gave ester 3. Reduction of
the ester group of 3 with LiBH4 gave alcohol 4 in a yield
of 70%. Acylation of 4 using Ac2O and pyridine afforded
amide 5. It is noteworthy that reduction of 3 with LiBH4
gave a high yield of 4, which was crucial for the total yield.
In the previous scheme, LiBH4 is probably inactivated by
carboxylic acids at C-26 and C-50, giving a low yield, while
in the present scheme, the carboxylic acids are protected by
(S)-(ꢀ)-phenylethylamine, giving a high yield.
CO2H
HO
a
O
OH
HO2C
O
S
O
Ph
1
HO
N
H
b
Ph
O
O
O
OH
N
O
S
H
3
Ph
Ph
O
OH
OH
c
R
1
5
1'
N
H
R
OH
N
H
3
OAc
4
2
2'
4
5
Scheme 1. Preparation of amide 5 from 1. Reagents and conditions: (a)
(S)-(ꢀ)-phenylethylamine, PyBOPꢁ, HOBt, Et3N, DMF, quant.; (b)
LiBH4, THF, 70%; (c) Ac2O, pyridine, CH2Cl2, quant.
The chirality at the C-3 position of 5 was assigned by com-
paring its retention time on HPLC with those of authentic
diastereomers, (3R)-5 and (3S)-5.5 Injection of 5 lg of 5
gave the chromatogram that was satisfactory for identifica-
tion. HPLC analysis showed that 5 was identical to the
(3R)-diastereomer, meaning that the absolute configura-
tion of the HMG group of 1 is (S), as proven previously.1
Figure 2. HPLC analysis of amide 5 derived from 1. (a) Chromatogram of
amide 5 derived from 0.1 mg of 1. (b) Chromatogram of authentic (3R)-
and (3S)-5. (c) Co-chromatogram of amide 5 (a) and the authentic
diastereomers (b).
A scaled down method was also successful. Ester 3 pre-
pared from 0.1 mg (0.16 lmol) of 1 was usable for the next

Y. Hattori et al. / Tetrahedron: Asymmetry 18 (2007) 1183–1186
1185
Using the refined method, we successfully determined the
absolute configuration of the HMG group from 0.3 mg
(0.46 lmol) of 2 (Scheme 2). The chirality of 5 at the C-3
position was established as (R) on the basis of HPLC anal-
ysis. Thus, the absolute configuration of the HMG group
of 2 was determined to be (S).
Et3N (31 lL, 220 lmol) in DMF (0.5 mL) were added
PyBOPꢁ (76 mg, 150 lmol) and HOBt (20 mg, 150 lmol)
at 0 ꢂC under Ar. The mixture was stirred for 30 min at
room temperature, and then the reaction was quenched
with dilute aqueous HCl. The organic materials were
extracted with ether and concentrated. The residue was
purified by ODS cartridge (10 g, Bond Elut C18; Varian)
column chromatography eluting with H2O, H2O–MeOH
(9:1, 8:2, 7:3, 6:4, 5:5, 4:6, 3:7, 2:8, and 1:9), and MeOH
(30 mL of each). Concentration of the material eluted with
CO2H
HO
MeOH gave
3 (31 mg, quant.) as a colorless oil.
20
1
O
½aꢁD ¼ ꢀ14:5 (c 2.15, CHCl3). H NMR d: 0.58 (3H, s),
0.86 (3H, s), 0.93 (3H, s), 0.94 (3H, d, J = 6.6 Hz), 0.98
(3H, s), 1.10 (3H, s), 1.16 (2H, m), 1.28–1.49 (6H, m),
1.29 (3H, d, J = 7.2 Hz), 1.32 (3H, s), 1.46 (3H, d,
J = 3.2 Hz), 1.49 (3H, d, J = 6.9 Hz), 1.59–1.65 (6H, m),
1.90–2.08 (9H, m), 2.42 (1H, d, J = 14.6 Hz), 2.53 (1H, d,
J = 14.6 Hz), 2.55 (1H, s), 2.61 (1H, m), 3.05 (1H, q,
J = 7.2 Hz), 3.96 (1H, d, J = 8.1 Hz), 4.72 (1H, s), 4.96
(1H, s), 4.98 (1H, s), 5.12 (2H, m), 5.87 (1H, d,
J = 8.1 Hz), 6.75 (1H, d, J = 8.1 Hz), 7.23–7.34 (10H, m);
13C NMR d: 16.0, 16.3, 17.8, 17.9, 18.7, 21.7, 21.8, 22.1,
23.3, 24.5, 25.9, 27.1, 27.7, 30.5, 31.3, 31.9, 32.7, 34.3,
36.1, 36.7, 36.7, 43.1, 44.8, 45.4, 47.1, 48.2, 48.4, 48.7,
49.6, 49.6, 70.2 (C · 2), 73.2, 78.7, 111.6, 126.0 (C · 2),
126.1 (C · 2), 127.3, 127.3, 128.6 (C · 2), 128.7 (C · 2),
132.7, 134.8, 143.3 (C · 2), 150.4, 170.2, 172.3, 172.7;
HRFABMS calcd for C53H77N2O6 [(M+H)+] 837.5782,
found 837.5845.
OH
HO2C
S
O
OH
2
Ph
O
OH
R
N
H
OAc
5
Scheme 2. Preparation of amide 5 from 2.
As far as we have investigated, normal phase chromatogra-
phy is appropriate for separating (3R)-5 and (3S)-5. If anal-
ysis by HPLC is performed under different conditions,
(3R)- and (3S)-diastereomers should be distinguished from
each other. For this purpose, it would be convenient to iso-
late small amounts of them and compare the chemical shift
values of the acetyl groups, 2.03 ppm for the former and
2.00 ppm for the latter, in the H NMR spectra.2
1
4.3. (3R)-1-[(S)-Phenylethyl]mevalonamide 4
To a solution of 3 (31 mg, 37 lmol) in THF (0.5 mL) was
added LiBH4 (180 lL, 370 lmol, 2.0 M in THF; Aldrich)
at 0 ꢂC under Ar. The mixture was stirred for 12 h at room
temperature, and then the reaction was quenched with
aqueous HCl. The organic materials were extracted with
EtOAc and concentrated. The residue was purified by
ODS cartridge (10 g, Bond Elut C18) column chromato-
graphy eluting with H2O, H2O–MeOH (9:1, 8:2, 7:3, 6:4,
5:5, 4:6, 3:7, 2:8, and 1:9), and MeOH (30 mL of each).
Concentration of the material eluted with H2O–MeOH
(6:4 and 5:5) gave 4 (6.4 mg, 70%) as a colorless oil.
3. Conclusion
In conclusion, we have refined the previous procedure to
achieve an improvement in the total yield of 5 derived from
a natural product with the HMG group. This method
would be useful for determining the absolute configuration
of the HMG group occurring in small amounts in natural
products.
18
1
½aꢁD ¼ ꢀ73:8 (c 3.40, CHCl3). H NMR d: 1.27 (3H, s,
CH3-3), 1.50 (3H, d, J = 6.9 Hz, H-20), 1.71, 1.78 (each
1H, m, H-4), 2.25, 2.52 (each 1H, d, J = 14.3 Hz, H-2),
3.81, 3.91 (each 1H, m, H-5), 5.15 (1H, t, J = 6.9 Hz, H-
10), 6.41 (1H, br s, NH), 7.25–7.36 (5H, m, C6H5-10); 13C
NMR d: 21.8 (C-20), 27.0 (CH3-3), 42.1 (C-4), 46.8 (C-2),
48.8 (C-10), 59.6 (C-5), 72.5 (C-3), 126.1 (C6H5), 127.5
(C6H5), 128.8 (C6H5), 143.0 (C6H5), 171.4 (C-1); EIMS
(probe) 70 eV m/z (%) 251 [M]+ (16), 206 (12), 120 (100),
106 (40), 105 (91); HREIMS calcd for C14H21NO3 (M+)
251.1521, found 251.1507.
4. Experimental
4.1. General
1H and 13C NMR spectra were measured with a Bruker
DRX 500 FT-NMR spectrometer in CDCl3 at 500 and
125 MHz, respectively. Chemical shifts were relative to
tetramethylsilane as an internal standard. Mass spectra
were obtained in a Jeol JMS 700 mass spectrometer. Opti-
cal rotations were determined with a Jasco DIP-1000
polarimeter.
4.4. (3R)-5-O-Acetyl-1-[(S)-phenylethyl]mevalonamide 5
4.2. (25S,30S)-(ꢀ)-12a-Hydroxy-3a-{30-hydroxy-30-methyl-
[(S)-phenylethyl]glutarylamide}-24-methyllanosta-8,24(31)-
dien-26-[(S)-phenylethyl]amide 3
To a solution of 4 (6.4 mg, 26 lmol) in CH2Cl2 (0.5 mL)
were added Ac2O (11 lL, 100 lmol) and pyridine (7.4 lL,
100 lmol). The mixture was stirred for 13 h, and then the
reaction was diluted with water. The organic materials
were extracted with EtOAc and concentrated to give 5
To a solution of 1 (23 mg, 37 lmol), (S)-(ꢀ)-phenylethyl-
amine [19 lL, 150 lmol, ee (GLC) 98%; Aldrich], and

1186
Y. Hattori et al. / Tetrahedron: Asymmetry 18 (2007) 1183–1186
18
(7.5 mg, quant.) as a colorless oil. ½aꢁD ¼ ꢀ59:0 (c 3.75,
a silica gel column (YMC A-004, 300 · 4.6 mm) eluting
with CH2Cl2-i-PrOH (20:1) at a flow rate of 1.0 mL/min
with detection at 254 nm.
CHCl3). For the other spectral data, see Refs. 2,8.
4.5. Preparation of amide 5 from 1 on a small scale
To a solution of 1 (0.1 mg, 0.16 lmol), (S)-(ꢀ)-phenylethyl-
amine (1 drop from a micro syringe), and Et3N (1 drop from
a micro syringe) in DMF (0.1 mL) were added PyBOPꢁ
(0.3 mg, 0.6 lmol) and HOBt (0.1 mg, 0.7 lmol) at 0 ꢂC
under Ar. The mixture was stirred for 1 h at room temper-
ature, and then the reaction was quenched with diluted
aqueous HCl. The organic materials were extracted with
ether. Evaporation of the solvent gave crude 3, which was
immediately used for the next step without purification.
To a solution of crude 3 in THF (0.1 mL) was added LiBH4
(1 drop from a micro syringe, 2.0 M in THF; Aldrich) at
0 ꢂC under Ar. The mixture was stirred for 12 h at room
temperature, and then the reaction was quenched with
aqueous HCl. The organic materials were extracted with
EtOAc. Evaporation of the solvent gave an oil, which was
purified by ODS cartridge (500 mg, Bond Elut C18) column
chromatography eluting with H2O, H2O–MeOH (9:1, 8:2,
7:3, 6:4, 5:5, 4:6, 3:7, 2:8, and 1:9), and MeOH (3 mL of
each). The material eluted with H2O–MeOH (8:2, 7:3, 6:4,
5:5, and 4:6) was concentrated and then subjected to HPLC
using an ODS column (Shodex C18-5B, 250 · 4.6 mm) elut-
ing with H2O–MeOH (1:1) at a flow rate of 1.0 mL/min
with detection at 212 nm. The material eluted at tR
8.4 min was separately collected and concentrated to give
4. To a solution of 4 in CH2Cl2 (0.1 mL) were added
Ac2O (1 drop from a micro syringe) and pyridine (2 drops
from a micro syringe). The mixture was stirred for 6 h,
and then the reaction was diluted with water. The organic
materials were extracted with EtOAc, and evaporation of
the solvent gave 5 (16 lg, 35% from 1).
References
1. Kamo, T.; Asanoma, M.; Shibata, H.; Hirota, M. J. Nat.
Prod. 2003, 66, 1104–1106.
2. Kamo, T.; Hirai, N.; Matsumoto, C.; Ohigashi, H.; Hirota,
M. Phytochemistry 2004, 65, 2517–2520.
3. Mimaki, Y.; Kawashima, K.; Kanmoto, T.; Sashida, Y.
Phytochemistry 1993, 34, 799–805.
4. Tanaka, M.; Hashimoto, K.; Okuno, T.; Shirahama, H.
Phytochemistry 1993, 34, 661–664.
5. Tanaka, M.; Hashimoto, K.; Okuno, T.; Shirahama, H.
Phytochemistry 1992, 31, 4355–4356.
6. Nozoe, S.; Takahashi, A.; Ohta, T. Chem. Pharm. Bull. 1993,
41, 1738–1742.
7. Hashimoto, K.; Kitaguchi, J.; Mizuno, Y.; Kobayashi, T.;
Shirahama, H. Tetrahedron Lett. 1996, 37, 2275–2278.
8. Kasai, R.; Miyakoshi, M.; Matsumoto, K.; Nie, R.-L.; Zhou,
J.; Morita, T.; Tanaka, O. Chem. Pharm. Bull. 1986, 34,
3974–3977.
9. Fujimoto, H.; Nakamura, E.; Kim, Y.-P.; Okuyama, E.;
Ishibashi, M.; Sassa, T. J. Nat. Prod. 2001, 64, 1234–1237.
10. Sassa, T.; Nukina, M. Agric. Biol. Chem. 1984, 48, 1923–
1925.
11. Kleinwa¨chter, P.; Luhmann, U.; Schlegel, B.; Heinze, S.;
Ha¨rtl, A.; Kiet, T. T.; Gra¨fe, U. J. Basic Microbiol. 1999, 39,
345–349.
12. Keller, A. C.; Maillard, M. P.; Hostettmann, K. Phytochem-
istry 1996, 41, 1041–1046.
13. Fujimoto, H.; Takano, Y.; Yamazaki, M. Chem. Pharm. Bull.
1992, 40, 869–872.
14. Bocchi, M.; Garlaschelli, L.; Vidari, G. J. Nat. Prod. 1992,
55, 428–431.
15. Takahashi, A.; Kusano, G.; Ohta, T.; Ohizumi, Y.; Nozoe, S.
Chem. Pharm. Bull. 1989, 37, 3247–3250.
4.6. Calibration curve of 5
16. Bernardi, M. D.; Fronza, G.; Gianotti, M. P.; Mellerio, G.;
Vidari, G.; Vita-Finzi, P. Tetrahedron Lett. 1983, 24, 1635–
1638.
17. Horie, T.; Tsukayama, M.; Yamada, T.; Miura, I.; Naka-
yama, M. Phytochemistry 1986, 25, 2621–2624.
18. Kumamoto, H.; Matsubara, Y.; Iizuka, Y.; Okamoto, K.;
Yokoi, K. Agric. Biol. Chem. 1985, 49, 2797–2798.
19. Baba, K.; Taniguti, M.; Yoneda, Y.; Kozawa, M. Phyto-
chemistry 1990, 29, 247–249.
20. Nishizawa, M.; Izuhara, R.; Kaneko, K.; Koshihara, Y.;
Fujimoto, Y. Chem. Pharm. Bull. 1988, 36, 87–95.
21. Nishizawa, M.; Fujimoto, Y. Chem. Pharm. Bull. 1986, 34,
1419–1421.
22. Tan, R. X.; Ma, W. G.; Wei, H. X.; Zhang, L. X.
Phytochemistry 1998, 48, 1245–1250.
A linear relationship between the amount of 5 and the peak
area was observed in the range of 5–250 ng by using an
ODS column (Shodex C18-5B 250 · 4.6 mm) eluting with
H2O–MeOH (1:1) at a flow rate of 1.0 mL/min with detec-
tion at 212 nm. The tR of 5 was 16.5 min.
4.7. HPLC analysis of (3R)-5 and (3S)-5
Amide 5 (5 lg, 17 nmol) in CH2Cl2-i-PrOH (20:1) was sub-
jected to HPLC with a silica gel column (YMC A-004,
300 · 4.6 mm) eluting with CH2Cl2-i-PrOH (20:1) at a flow
rate of 1.0 mL/min with detection at 254 nm. The tRs of
(3R)-5 and (3S)-5 were 9.8 and 10.5 min, respectively.
23. Yuda, M.; Ohtani, K.; Mizutani, K.; Kasai, R.; Tanaka, O.;
Jia, M.-R.; Ling, Y.-R.; Pu, X.-F.; Saruwatari, Y. Phyto-
chemistry 1990, 29, 1989–1990.
24. Mizutani, K.; Yuda, M.; Tanaka, O.; Saruwatari, Y.; Jia,
M.-R.; Ling, Y.-K.; Pu, X.-F. Chem. Pharm. Bull. 1988, 36,
2726–2729.
4.8. Determination of the absolute configuration of 2
Amide 5 (24 lg, 80 nmol) was prepared from 2 (0.3 mg,
0.46 lmol) by the same procedure as described in Section
4.5. Amide 5 (5 lg, 17 nmol) was subjected to HPLC with
25. Berget, B. J.; Baker, F. C.; Lee, E.; Schooley, D. A. J. Am.
Chem. Soc. 1979, 101, 7432–7434.