Notes
J ournal of Natural Products, 2002, Vol. 65, No. 9 1365
over silica gel by elution with CH2Cl2-MeOH (95:5) to give
three subfractions (DG1-DG3). Subfraction DG2 (802.7 mg)
was further purified by column chromatography over LiChro-
prep RP-18 (2.5 × 30 cm) using 30% CH3CN as eluting solvent
and finally purified by preparative HPLC eluted with hexane-
MeOH-i-PrOH (70:17:13) to afford 640.1 mg of compound 1
and 84.2 mg of compound 2.
Fraction DH (2.3 g) was purified by column chromatography
over silica gel using CH2Cl2-MeOH (90:10) to give six sub-
fractions (DH1-DH6). Subfraction DH3 (382.6 mg) was fur-
ther purified by column chromatography over LiChroprep RP-
18 (2.5 × 30 cm) using 30% CH3CN for elution and finally
purified by preparative HPLC as described above to give 5.0
mg of compound 3.
The EtOAc-soluble layer (2.93 g) was chromatographed on
silica gel using gradient mixtures of CH2Cl2-MeOH (95:5 f
50:50) to give seven fractions (EA-EG). Fraction EC (165.2
mg) was purified by column chromatography over silica gel
with CH2Cl2-MeOH (90:10) to give three fractions (EC1-
EC3). Fraction EC2 was further purified by HPLC using CH3-
CN-H2O (25:75) to give an additional amount of compound 3
(3 mg). Fraction EE (501.4 mg) was purified by column
chromatography over silica gel with EtOAc-MeOH (90:10 f
25:75) to give six fractions (EE1-EE6). Fraction EE2 was
purified by preparative TLC on silica gel (1 mm, 20 × 20 cm,
Merck), developed with CH2Cl2-MeOH (80:20), to afford 15.9
mg of compound 4.
resonances due to an iridolactone moiety. The chemical
shifts of all carbons were determined by correlation with
the proton chemical shifts assigned by detailed decoupling
experiments. Accordingly, the 13C NMR spectrum of 2 was
fully assigned, as shown in Table 2. These data also
supported the proposed structure 2. The stereochemistry
of compound 2 was determined by NOESY data, which
showed correlations between H-9′′ and H-5′′ but not
between H-4′′ and H-5′′, demonstrating a cis configuration
for the C-11′′ carbonyl group and H-5′′ (Figure 2). As a
result, the absolute structure of compound 2 was deter-
mined as shown.
Compound 3 was obtained as an amorphous powder,
whose composition was determined to be C28H43O14 by
HRFABMS. Compound 3 was considered to be closely
related structurally to compound 1, since their IR and UV
spectra were essentially the same. Compound 3 exhibited
1
two distinctive signals for methoxy groups from the H and
13C NMR spectra. An upfield shift of the H-3′′ signal was
observed, indicating the opening of the lactone ring of 1
(Table 1). The configuration of the methyl ester could be
proposed, since a connectivity in the HMBC spectrum was
observed between H-9′′ at δ 2.55 and C-1′′ at δ 177.4 of
the methyl ester. The correlations between H-3′′ (δ 3.72
and 3.83) and C-4′′ (δ 51.7), and H-3′′ and C-11′′ (δ 175.4),
were also observed. The structure of 3 was thus estab-
lished. Since compounds 1 and 2 did not decompose under
any of the laboratory conditions applied or in solution of
MeOH, compound 3 does not appear to be an extraction
artifact.
Asp er u loid e A (1): colorless needles from MeOH-CH2Cl2
(9:1); [R]22 -19.5° (c 1.52, CHCl3); UV λmax (MeOH) 235 (log
D
ꢀ 3.93) nm; IR (KBr) νmax 3422, 2957, 1718, 1646 cm-1; 1H and
13C NMR data, see Tables 1 and 2; HRFABMS m/z 571.2382
[M + H]+ (calcd for C27H39O13, 571.2391).
X-r a y Str u ctu r e Deter m in a tion of 1.8 Suitable colorless
crystals of 1 were obtained by recrystallization from MeOH-
CH2Cl2 (9:1). The X-ray data were collected on an Enraf-
Nonius CAD4 automated diffractometer equipped with a Mo
X-ray tube (λ ) 0.71073 Å) and a graphite crystal monochro-
mator. The crystal (0.2 × 0.1 × 0.05 mm) belongs to the
monoclinic system, space group C2, with a ) 26.144(7) Å, b )
Picconioside II was identified by comparing its physical
and spectral data with literature values.2
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. Melting points were
determined on a Thomas-Hoover capillary melting apparatus
and are uncorrected. Optical rotations were determined on an
Autopol III automatic polarimeter (Rudolph Research Co.,
Flanders, NJ ). UV spectra were taken with a Shimazu UV 240
UV-visible recording spectrometer. IR spectra were recorded
on a Perkin-Elmer 16F-PC FT-IR instrument using potassium
bromide pellets. NMR spectra were recorded on a Bruker
AMX-500 (500 MHz) spectrometer. 1H-1H COSY, NOE,
NOESY, HMQC, and HMBC NMR spectra were obtained with
the usual pulse sequences, and data processing was performed
with the standard Bruker software. HRFABMS were deter-
mined on a J EOL J MS-HX 110/100A mass spectrometer.
Preparative HPLC was performed on a Waters pump (model
510) with a photodiode array detector (Waters model 996)
using a ChiraSper (10 mm × 250 mm, Merck) column. TLC
and column chromatography were carried out on precoated
silica gel F254 plates (Merck, art. 5715), RP-18 F254S plates
(Merck, art. 15423), silica gel 60 (Merck, 230-400 mesh), and
Lichroprep RP-18 (Merck, 40-63 µm).
P la n t Ma ter ia l. The whole plants of Asperula maximow-
iczii were collected at Mt. Odae, Korea, in August 1994, and
the species was identified by Dr. J ong Hwan Kwak at Korea
Institute of Science and Technology (KIST). Voucher specimens
(398-20) have been deposited in our laboratory at KIST.
Extr a ction a n d Isola tion . The whole plants (697.0 g) were
air-dried in the dark and then extracted with a methanol to
give methanol-soluble extract. The dried extract residue (97.0
g) was suspended in water and sequentially partitioned with
methylene chloride, ethyl acetate, and n-butanol. The meth-
ylene chloride extract was evaporated under reduced pressure
to yield 30.4 g of a residue. This residue was divided into 12
fractions (DA-DL) by column chromatography on silica gel
using the following solvent mixtures: CH2Cl2-EtOAc (75:25
f 25:75), CH2Cl2-MeOH (90:10 f 25:75), and a MeOH wash.
Fraction DG (3.7 g) was purified by column chromatography
8.704(5) Å, c ) 14.209(6) Å, V ) 3131(2) Å3, Z ) 4, Dcalcd
)
1.342 g/cm3. The orientation matrix and unit cell dimensions
were determined from 25 machine-centered reflections in the
2θ range from 15° to 25°. The variation of intensities was
monitored by a repeated check of intensities of three reflections
every 1 h during the data collection period. A direct method
(SHELXS-90)9 was employed to find all the atoms. Subsequent
cycles of Fourier map and least-squares refinement were
followed (SHELXL-97).9 Absolute configuration was deter-
mined by known conformation of the anomeric proton of â-D-
glucose of the loganin moiety. All non-hydrogen atoms were
refined anisotropically for 1. Hydrogen atoms were included
in the structure factor calculation using a riding model. All
the calculations were carried out using VAX and PC comput-
ers. The refinement converged to a final R1 ) 0.083, wR2
)
0.2238, where R1 ) ∑||Fo| - |Fc||/∑|Fo, wR2 ) {∑w(Fo - Fc2)2/
2
4
∑wFo
}
1/2, for 1860 observed reflections [I > 2σ(I)] and 397
variable parameters.
Asp er u loid e B (2): amorphous powder; [R]18 -40.6° (c
D
0.38, MeOH); UV λmax (MeOH) 236 (log ꢀ 4.02) nm; IR (KBr)
νmax 3414, 2956, 1726, 1634, 1440, 1288, 1076 cm-1; H and
1
13C NMR data, see Tables 1 and 2; HRFABMS m/z 571.2410
[M + H]+ (calcd for C27H39O13, 571.2391).
Asp er u loid e C (3): amorphous powder; [R]19 -53.2° (c
D
0.063, MeOH); UV λmax (MeOH) 235 (log ꢀ 3.88) nm; IR (KBr)
νmax 3422, 2956, 1718, 1634, 1440, 1288, 1076 cm-1; H and
1
13C NMR data, see Tables 1 and 2; HRFABMS m/z 603.2663
[M + H]+ (calcd for C28H43O14, 603.2653).
Alk a lin e Hyd r olysis of Asp er u loid e A (1). To a solution
of compound 1 (53.2 mg) in MeOH (3 mL) was added 1 N
NaOH (2 mL), and the mixture was stirred at room temper-
ature for 12 h. The reaction mixture was neutralized by
addition of Amberlite IR-120 (H+-form). After removing the
resin by filtration, the filtrate was evaporated. The residue
was dissolved in 2 mL of MeOH and treated with CH2N2. After
the reaction was completed, the solvent was evaporated in