890
H.-H. Xie and M. Yoshikawa
with Dowex HCR W2 resin (Hþ form),
and filtrated to give a hydrolysate.
was added to 1 M HCl (0.5 ml) and stirred
at 808C for 3 h. The solution was extracted
with EtOAc (0.5 ml £ 3) and the aqueous
layer was neutralized with Amberlite IRA-
400 resin (OH– form) and then filtrated.
The solution was condensed to ca. 0.15 ml
and passed through a filter (0.45 mm),
and then subjected to HPLC analysis
[column: Kaseisorb LC NH2-60-5 (5 mm,
250 £ 4.6 mm i.d.; Tokyo Kasei Co.,
Tokyo, Japan); detection: Shodex OR-2
optical rotation detector; mobile phase:
85% aqueous CH3CN (v/v); flow rate:
0.8 ml/min] to identify D-glucose and
L-rhamnose by comparison of the retention
time and optical rotation with authentic
samples (L-rhamnose: tR 7.8 min, negative
optical rotation; D-glucose: tR 13.9 min,
positive optical rotation) [3,4].
3.4.1 Characterization of 1a, 2a, and 4a
A part of the hydrolysate (1–3: 0.6 ml, 4:
0.4 ml) was condensed under reduced
pressure to give a residue, which was
dissolved in MeOH (0.1 ml) and passed
through a filter (0.45 mm) and then
subjected to HPLC analysis to characterize
kaempferol 3-O-b-D-glucopyranosyl-7-
O-a-L-rhamnopyranoside (1a) from 1,
quercetin 3-O-b-D-glucopyranosyl-7-O-
a-L-rhamnopyranoside (2a) from 2 and 3,
and sinocrassoside D2 (4a) from 4 by
direct comparison with the samples
previously obtained from this herb [3,5].
3.4.2 Characterization of (S)-(þ)-2-
methylbutanoic acid
A part of the hydrolysate of 4 (0.4 ml) was
condensed under vacuum to yield a
residue, which was dissolved in dichlor-
oethane (1 ml) and added p-nitrobenzyl-
N,N0-diisopropylisourea (5 mg) and stirred
at 808C for 1 h. Evaporation of the reaction
mixture gave a residue, which was
dissolved in MeOH (0.1 ml) and passed
through a filter (0.45 mm), and then
subjected to HPLC analysis [column:
YMC-Pack ODS-AQ (5 mm, 250 £
4.6 mm i.d.; YMC Co., Kyoto, Japan);
mobile phase: 20% CH3CN in 0.1%
aqueous H3PO4 (v/v); detection: Shodex
OR-2 optical rotation detector (Showa
Denko Co., Tokyo, Japan); flow rate:
1.0 ml/min] to identify S-(þ)-2-methylbu-
tyric acid (tR: 15.4 min, positive optical
rotation) by direct comparison of its
retention time and optical rotation with
the commercial standard [2,3].
Acknowledgements
The research was financially supported by the
21st Centre of Excellence Program, Academic
Frontier Project from the Ministry of Edu-
cation, Culture, Sports, Science and Technol-
ogy of Japan and Guangzhou Science and
Technology Project (Grant No. 12C14061591).
References
[1] State Administration of Traditional Chi-
nese Medicine, Chinese Materia Medica
(Shanghai Science & Technology Press,
Shanghai, China, 1999), Vol. 9, p. 778.
[2] M. Yoshikawa, T. Wang, T. Morikawa, H.
Xie, and H. Matsuda, Chem. Pharm. Bull.
55, 1308 (2007).
[3] T. Morikawa, H. Xie, T. Wang, H.
Matsuda, and M. Yoshikawa, Chem.
Pharm. Bull. 56, 1438 (2008).
[4] K. Ninomiya, T. Morikawa, H. Xie, H.
Matsuda, and M. Yoshikawa, Heterocycles
75, 1983 (2008).
[5] T. Morikawa, H. Xie, T. Wang, H.
Matsuda, and M. Yoshikawa, Chem.
Biodiver. 6, 411 (2009).
[6] H. Xie and M. Yoshikawa, J. Asian Nat.
Prod. Res. 14, 503 (2012).
[7] H. Xie, T. Morikawa, H. Matsuda, S.
Nakamura, O. Muraoka, and M. Yoshikawa,
Chem. Pharm. Bull. 54, 669 (2006).
3.4.3 Characterization of D-glucose and
L-rhamnose
The rest of the hydrolysate was condensed
under vacuum to afford a residue, which